WO2024001652A1 - Network adaptation method and apparatus between devices, storage medium, and electronic device - Google Patents

Abstract

A network adaptation method and apparatus between devices, a storage medium, and an electronic device, relating to the technical field of network communications. The method comprises: obtaining, from a kernel node of a physical port associated with a master control device, a network node of a slave device accessing the master control device; reading, from the network node, the interface number of a universal serial bus interface of the master control device via which the slave device accesses the master control device; and generating a communication address between the master control device and the slave device on the basis of the interface number and the current communication network segment of the slave device.

Description

设备间的网络适配方法及装置、存储介质、电子设备Network adaptation method and device between devices, storage media, electronic equipment

相关申请的交叉引用Cross-references to related applications

本公开以申请号为：202210762603.X，申请日为：2022年06月29日，发明名称为：设备间的网络适配方法及装置、存储介质、电子设备的申请文件作为优先权，该中国专利申请的全部内容通过引用全部并入本文。This disclosure takes priority as the application document with the application number: 202210762603. The entire contents of the patent application are incorporated herein by reference.

技术领域Technical field

本公开实施例涉及网络通信技术领域，具体而言，涉及一种设备间的网络适配方法、设备间的网络适配装置、计算机可读存储介质以及电子设备。Embodiments of the present disclosure relate to the field of network communication technology, and specifically, to a network adaptation method between devices, a network adaptation device between devices, a computer-readable storage medium, and an electronic device.

背景技术Background technique

现有的方法中，主控设备和从属设备之间存在通信地址不确定的问题，进而导致主控设备和从属设备之间的通信效率较低。In the existing method, there is a problem of uncertain communication addresses between the master control device and the slave devices, which leads to low communication efficiency between the master control device and the slave devices.

但是，由于微裂纹前期不会造成短路，进而使得所检测到的裂纹缺陷的精确度较低。However, since microcracks do not cause short circuits in the early stage, the accuracy of detected crack defects is low.

发明内容Contents of the invention

根据本公开的一个方面，提供一种设备间的网络适配方法，配置于终端设备，所述终端设备中包括主控设备以及从属设备；所述设备间的网络适配方法包括：According to one aspect of the present disclosure, a network adaptation method between devices is provided, which is configured in a terminal device. The terminal device includes a master device and a slave device; the network adaptation method between devices includes:

从与所述主控设备关联的物理端口的内核节点中获取接入的从属设备的网络节点；Obtain the network node of the accessed slave device from the kernel node of the physical port associated with the master device;

从所述网络节点中读取所述从属设备在接入所述主控设备时，所经由的主控设备的通用串行总线接口的接口编号；Read from the network node the interface number of the universal serial bus interface of the master device through which the slave device accesses the master device;

基于所述接口编号以及所述从属设备所具有的当前通信网段，生成所述主控设备与所述从属设备之间的通信地址。Based on the interface number and the current communication network segment of the slave device, a communication address between the master device and the slave device is generated.

在本公开的一种示例性实施例中，在从与所述主控设备关联的物理端口的目录下，获取接入的从属设备的网络节点之前，所述设备间的网络适配方法还包括：In an exemplary embodiment of the present disclosure, before obtaining the network node of the accessed slave device from the directory of the physical port associated with the master control device, the network adaptation method between devices further includes :

启动预设程序进程，并基于所述预设程序进程对所述主控设备的物理端口的内核节点进行轮询监测；Start a preset program process, and perform polling monitoring of the kernel node of the physical port of the main control device based on the preset program process;

在监测到所述物理端口中有从属设备接入时，在所述内核节点中创建与接入的从属设备关联的网络节点。When it is detected that a slave device is connected to the physical port, a network node associated with the connected slave device is created in the kernel node.

在本公开的一种示例性实施例中，所述设备间的网络适配方法还包括：In an exemplary embodiment of the present disclosure, the network adaptation method between devices further includes:

在所述主控设备中建立与接入的从属设备关联的虚拟节点，并计算已经建立的虚拟节点的节点数量； Establish a virtual node associated with the accessed slave device in the master control device, and calculate the number of established virtual nodes;

判断所述节点数量是否大于所述主控设备所具有的通用串行总线接口的接口数量；Determine whether the number of nodes is greater than the number of universal serial bus interfaces of the master control device;

在确定所述节点数量大于等于所述接口数量时，终止所述从属设备的接入。When it is determined that the number of nodes is greater than or equal to the number of interfaces, access of the slave device is terminated.

在本公开的一种示例性实施例中，所述设备间的网络适配方法还包括：In an exemplary embodiment of the present disclosure, the network adaptation method between devices further includes:

根据所述通信地址建立所述主控设备与从属设备之间的通信连接，并基于所述通信连接，在所述主控设备以及从属设备之间进行数据传输和/或指令传输。A communication connection is established between the master device and the slave device according to the communication address, and based on the communication connection, data transmission and/or instruction transmission is performed between the master device and the slave device.

在本公开的一种示例性实施例中，所述设备间的网络适配方法还包括：In an exemplary embodiment of the present disclosure, the network adaptation method between devices further includes:

根据所述通信地址，建立所述主控设备的通用串行总线接口与所述从属设备之间的当前设备映射关系表；According to the communication address, establish a current device mapping relationship table between the universal serial bus interface of the master device and the slave device;

判断所述当前设备映射关系表与历史设备映射关系表之间是否存在新接入的从属设备；Determine whether there is a newly accessed slave device between the current device mapping relationship table and the historical device mapping relationship table;

若存在新接入的从属设备，则为新接入的从属设备增加路由路径，若不存在新接入的从属设备，则对所述历史设备映射关系表中所包括的所有路由路径进行刷新。If there is a newly accessed slave device, a routing path is added for the newly accessed slave device. If there is no newly accessed slave device, all routing paths included in the historical device mapping relationship table are refreshed.

在本公开的一种示例性实施例中，所述设备间的网络适配方法还包括：In an exemplary embodiment of the present disclosure, the network adaptation method between devices further includes:

根据所述路由路径建立所述主控设备和/或从属设备与外部应用程序之间的通信连接。A communication connection between the master device and/or the slave device and an external application is established according to the routing path.

在本公开的一种示例性实施例中，所述终端设备包括智能冰箱，所述主控设备包括主控SoC板卡，所述从属设备包括一个或者多个智能摄像头；In an exemplary embodiment of the present disclosure, the terminal device includes a smart refrigerator, the main control device includes a main control SoC board, and the slave device includes one or more smart cameras;

其中，基于所述通信连接，在所述主控设备以及从属设备之间进行数据传输和/或指令传输，包括：Wherein, based on the communication connection, data transmission and/or instruction transmission is performed between the master control device and the slave device, including:

根据所述从属设备在所述终端设备中所处的当前位置，在所述从属设备中部署计算机视觉算法；Deploying a computer vision algorithm in the slave device according to the current location of the slave device in the terminal device;

控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，并基于所述通信连接图像识别结果从从属设备传输至主控设备。The slave device is controlled to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result, and the image recognition result is transmitted from the slave device to the master control device based on the communication connection.

在本公开的一种示例性实施例中，所述计算机视觉算法包括人像识别算法和/或物品识别算法；所述当前位置包括智能冰箱内部和/或智能冰箱外部；In an exemplary embodiment of the present disclosure, the computer vision algorithm includes a portrait recognition algorithm and/or an object recognition algorithm; the current location includes the inside of a smart refrigerator and/or the outside of a smart refrigerator;

其中，根据所述从属设备在所述终端设备中所处的当前位置，在所述从属设备中部署计算机视觉算法，包括：Wherein, deploying a computer vision algorithm in the slave device according to the current location of the slave device in the terminal device includes:

当所述智能摄像头所处的当前位置位于智能冰箱内部时，在该智能摄像头中部署物品识别算法；When the current location of the smart camera is inside the smart refrigerator, deploy an object recognition algorithm in the smart camera;

当所述智能摄像头所处的当前位置位于智能冰箱内部时，在该智能摄像头中部署人像识别算法。When the current location of the smart camera is inside the smart refrigerator, a portrait recognition algorithm is deployed in the smart camera.

在本公开的一种示例性实施例中，控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，包括：In an exemplary embodiment of the present disclosure, controlling the slave device to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result includes:

主控设备在检测到与所述智能冰箱对应的第一预设位置存在目标对象时，根据主控设备与从属设备之间的当前设备映射关系表，获取处于该第一预设位置的从属设备与主控设备之间的通信地址； When the master control device detects that a target object exists at the first preset position corresponding to the smart refrigerator, it obtains the slave device at the first preset position according to the current device mapping relationship table between the master control device and the slave device. Communication address with the main control device;

主控设备生成与处于第一预设位置的从属设备关联的人脸图像采集指令，并基于处于第一预设位置的从属设备与主控设备之间的通信地址，将所述人脸图像采集指令发送至处于第一预设位置的从属设备；The master control device generates a face image collection instruction associated with the slave device at the first preset position, and collects the face image based on the communication address between the slave device at the first preset position and the master control device. sending the command to the slave device in the first preset position;

处于第一预设位置的从属设备响应于所述人脸图像采集指令，采集所述目标对象的待识别人脸图像，并利用预设的人像识别算法对采集到的待识别人脸图像进行识别，得到人像识别结果。The slave device at the first preset position responds to the face image collection instruction, collects the face image to be recognized of the target object, and uses a preset portrait recognition algorithm to identify the collected face image to be recognized. , get the portrait recognition result.

在本公开的一种示例性实施例中，利用预设的人像识别算法对采集到的待识别人脸图像进行识别，得到人像识别结果，包括：In an exemplary embodiment of the present disclosure, a preset portrait recognition algorithm is used to recognize the collected face image to be recognized, and a portrait recognition result is obtained, including:

利用预设的人脸检测与关键点定位工具，检测所述待识别人脸图像的待识别人脸区域；Use preset face detection and key point positioning tools to detect the face area to be recognized in the face image to be recognized;

在所述待识别人脸区域中提取所述待识别人脸图像的待识别脸部关键点，并根据所述待识别脸部关键点，计算所述待识别人脸特征；Extract the face key points to be recognized of the face image to be recognized in the face area to be recognized, and calculate the face features to be recognized based on the face key points to be recognized;

计算所述待识别人脸特征以及处于第二预设位置处的从属设备中的特征值库中的原始人脸特征之间的欧式距离；Calculate the Euclidean distance between the facial feature to be recognized and the original facial feature in the feature value library in the slave device at the second preset position;

根据所述欧式距离，得到所述待识别人脸特征与所述原始人脸特征之间的相似度比对结果，并根据所述相似度比对结果得到所述人像识别结果。According to the Euclidean distance, a similarity comparison result between the facial feature to be recognized and the original facial feature is obtained, and the portrait recognition result is obtained based on the similarity comparison result.

在本公开的一种示例性实施例中，控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，包括：In an exemplary embodiment of the present disclosure, controlling the slave device to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result includes:

主控设备在检测到与所述智能冰箱对应的第二预设位置存在新增物品时，根据主控设备与从属设备之间的当前设备映射关系表，获取处于该第二预设位置的从属设备与主控设备之间的通信地址；When the master control device detects that there is a new item in the second preset position corresponding to the smart refrigerator, it obtains the slave at the second preset position according to the current device mapping relationship table between the master control device and the slave device. Communication address between the device and the main control device;

主控设备生成与处于第二预设位置的从属设备关联的物品图像采集指令，并基于处于第二预设位置的从属设备与主控设备之间的通信地址，将所述物品图像采集指令发送至处于第二预设位置的从属设备；The master control device generates an item image collection instruction associated with the slave device at the second preset position, and sends the item image collection instruction based on the communication address between the slave device at the second preset position and the master control device. to the slave device in the second preset position;

处于第二预设位置的从属设备响应于所述物品图像采集指令，采集所述新增物品的待识别物品图像，并利用预设的物品识别算法对采集到的待识别物品图像进行识别，得到物品图像识别结果。The slave device at the second preset position responds to the item image collection instruction, collects the item image to be identified of the newly added item, and uses the preset item recognition algorithm to identify the collected item image to be identified, and obtains Item image recognition results.

根据本公开的一个方面，提供一种设备间的网络适配装置，配置于终端设备，所述终端设备中包括主控设备以及从属设备；所述设备间的网络适配装置包括：According to one aspect of the present disclosure, a network adaptation device between devices is provided, which is configured in a terminal device. The terminal device includes a master control device and a slave device; the network adaptation device between devices includes:

网络节点获取模块，用于从与所述主控设备关联的物理端口的内核节点中获取接入的从属设备的网络节点；A network node acquisition module, configured to acquire the network node of the accessed slave device from the kernel node of the physical port associated with the master control device;

接口编号读取模块，用于从所述网络节点中读取所述从属设备在接入所述主控设备时，所经由的主控设备的通用串行总线接口的接口编号；An interface number reading module, configured to read from the network node the interface number of the universal serial bus interface of the master device through which the slave device accesses the master device;

通信地址生成模块，用于基于所述接口编号以及所述从属设备所具有的当前通信网段，生成所述主控设备与所述从属设备之间的通信地址。A communication address generation module, configured to generate a communication address between the master device and the slave device based on the interface number and the current communication network segment of the slave device.

根据本公开的一个方面，提供一种计算机可读存储介质，其上存储有计算机程序，所 述计算机程序被处理器执行时实现上述任意一项所述的设备间的网络适配方法。According to an aspect of the present disclosure, a computer-readable storage medium is provided with a computer program stored thereon, so When the computer program is executed by the processor, the network adaptation method between devices described in any one of the above is implemented.

根据本公开的一个方面，提供一种电子设备，包括：According to an aspect of the present disclosure, an electronic device is provided, including:

处理器；以及processor; and

存储器，用于存储所述处理器的可执行指令；memory for storing executable instructions for the processor;

其中，所述处理器配置为经由执行所述可执行指令来执行上述任意一项所述的设备间的网络适配方法。Wherein, the processor is configured to execute the network adaptation method between devices described in any one of the above by executing the executable instructions.

应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，并不能限制本公开。It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only, and do not limit the present disclosure.

附图说明Description of drawings

此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本公开的实施例，并与说明书一起用于解释本公开的原理。显而易见地，下面描述中的附图仅仅是本公开的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。在附图中：The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure. Obviously, the drawings in the following description are only some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting creative efforts. In the attached picture:

图1示意性示出根据本公开示例实施例的一种设备间的网络适配方法的流程图。FIG. 1 schematically shows a flow chart of a network adaptation method between devices according to an exemplary embodiment of the present disclosure.

图2示意性示出根据本公开示例实施例的一种设备间的网络适配方法的应用场景示例图。FIG. 2 schematically shows an example application scenario diagram of a network adaptation method between devices according to an exemplary embodiment of the present disclosure.

图3示意性示出根据本公开示例实施例的一种主控设备的结构示例图。FIG. 3 schematically shows an example structural diagram of a main control device according to an exemplary embodiment of the present disclosure.

图4示意性示出根据本公开示例实施例的一种多个不同网段的主控设备以及从属设备的场景示例图。FIG. 4 schematically shows an example diagram of a scenario of master devices and slave devices in multiple different network segments according to an exemplary embodiment of the present disclosure.

图5示意性示出根据本公开示例实施例的一种路由路径更新的方法流程图。FIG. 5 schematically shows a flow chart of a routing path update method according to an example embodiment of the present disclosure.

图6示意性示出根据本公开示例实施例的一种智能冰箱的示例图。FIG. 6 schematically shows an example diagram of a smart refrigerator according to an example embodiment of the present disclosure.

图7示意性示出根据本公开示例实施例的一种YOLOV4的结构示例图。Figure 7 schematically shows an example structural diagram of a YOLOV4 according to an example embodiment of the present disclosure.

图8示意性示出根据本公开示例实施例的一种主干特征提取网络(BackBone)的结构示例图。FIG. 8 schematically shows an example structural diagram of a backbone feature extraction network (BackBone) according to an exemplary embodiment of the present disclosure.

图9示意性示出根据本公开示例实施例的一种颈部特征融合网络(Neck)的结构示例图。Figure 9 schematically shows an example structural diagram of a neck feature fusion network (Neck) according to an example embodiment of the present disclosure.

图10示意性示出根据本公开示例实施例的一种头部特征检测网络(Head)的结构示例图。Figure 10 schematically shows an example structural diagram of a head feature detection network (Head) according to an exemplary embodiment of the present disclosure.

图11示意性示出根据本公开示例实施例的一种设备间的网络适配装置的框图。Figure 11 schematically shows a block diagram of a network adaptation device between devices according to an exemplary embodiment of the present disclosure.

图12示意性示出根据本公开示例实施例的一种用于实现上述设备间的网络适配方法的电子设备。FIG. 12 schematically illustrates an electronic device for implementing the above network adaptation method between devices according to an example embodiment of the present disclosure.

具体实施方式 Detailed ways

现在将参照附图更全面地描述示例实施方式。然而，示例实施方式能够以多种形式实施，且不应被理解为限于在此阐述的范例；相反，提供这些实施方式使得本公开将更加全面和完整，并将示例实施方式的构思全面地传达给本领域的技术人员。所描述的特征、结构或特性可以以任何合适的方式结合在一个或更多实施方式中。Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concepts of the example embodiments. To those skilled in the art. The described features, structures or characteristics may be combined in any suitable manner in one or more embodiments.

此外，附图仅为本公开的示意性图解，并非一定是按比例绘制。图中相同的附图标记表示相同或类似的部分，因而将省略对它们的重复描述。附图中所示的一些方框图是功能实体，不一定必须与物理或逻辑上独立的实体相对应。可以采用软件形式来实现这些功能实体，或在一个或多个硬件模块或集成电路中实现这些功能实体，或在不同网络和/或处理器装置和/或微控制器装置中实现这些功能实体。Furthermore, the drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and/or processor devices and/or microcontroller devices.

本示例实施方式中首先提供了一种设备间的网络适配方法，该方法可以运行于终端设备；其中，本申请所记载的终端设备，可以指用户设备(User Equipment，UE)、接入终端、V2X通信中的终端、用户单元、用户站、移动站、移动台、远方站、远程终端、移动设备、用户终端、终端设备、无线通信设备、用户代理或用户装置。终端还可以是蜂窝电话、无绳电话、会话启动协议(Session Initiation Protocol，SIP)电话、无线本地环路(Wireless Local Loop，WLL)站、个人数字处理(Personal Digital Assistant，PDA)、具有无线通信功能的手持设备、计算设备或连接到无线调制解调器的其它处理设备、车载设备、可穿戴设备，未来5G网络中的终端设备或者未来演进的公用陆地移动通信网络(Public Land Mobile Network，PLMN)中的终端设备等，本申请实施例对此并不限定。终端还可以包括V2X设备，例如为车辆或车辆中的车载单元(On Board Unit，OBU)；当然，本领域技术人员也可以根据需求在其他平台运行本公开的方法，本示例性实施例中对此不做特殊限定。This example implementation first provides a network adaptation method between devices, which can be run on terminal devices; wherein, the terminal devices recorded in this application can refer to user equipment (User Equipment, UE), access terminals , terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal equipment, wireless communication equipment, user agent or user device in V2X communication. The terminal can also be a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA), or a phone with wireless communication functions Handheld devices, computing devices or other processing devices connected to wireless modems, vehicle-mounted devices, wearable devices, terminal devices in future 5G networks or terminals in future evolved Public Land Mobile Communications Networks (Public Land Mobile Network, PLMN) Equipment, etc., the embodiments of this application are not limited to this. The terminal may also include a V2X device, such as a vehicle or an On Board Unit (OBU) in the vehicle; of course, those skilled in the art may also run the method of the present disclosure on other platforms according to requirements. In this exemplary embodiment, There are no special restrictions on this.

进一步的，参考图1所示，该设备间的网络适配方法可以包括以下步骤：Further, as shown in Figure 1, the network adaptation method between devices may include the following steps:

步骤S110.从与所述主控设备关联的物理端口的内核节点中获取接入的从属设备的网络节点；Step S110. Obtain the network node of the accessed slave device from the kernel node of the physical port associated with the master device;

步骤S120.从所述网络节点中读取所述从属设备在接入所述主控设备时，所经由的主控设备的通用串行总线接口的接口编号；Step S120. Read the interface number of the universal serial bus interface of the master device through which the slave device passes when accessing the master device from the network node;

步骤S130.基于所述接口编号以及所述从属设备所具有的当前通信网段，生成所述主控设备与所述从属设备之间的通信地址。Step S130. Generate a communication address between the master device and the slave device based on the interface number and the current communication network segment of the slave device.

上述设备间的网络适配方法中，一方面，由于在生成主控设备与从属设备之间的通信地址时，是基于从属设备在接入主控设备时所经由的通用串行总线接口的接口编号来生成的，因此，在从属设备未插拔出主控设备时，可以确保主控设备与从属设备之间的通信地址的唯一性，进而可以解决现有技术中由于主控设备和从属设备之间存在通信地址不确定，进而导致主控设备和从属设备之间的通信效率较低的问题；再一方面，可以通过从与主控设备关联的物理端口的内核节点中获取接入的从属设备的网络节点；然后从网络节点中读取从属设备在接入主控设备时，所经由的主控设备的通用串行总线接口的接口编号；最后基于接口编号以及从属设备所具有的当前通信网段，生成主控设备与从属设备之间的通信 地址；因此，当有新的从属设备接入主控设备时，可以直接获取对应的网络节点进而基于该网络节点生成对应的通信地址，进而使得从属设备与主控设备之间的通信地址不受从属设备接入或者挂载顺序的影响。In the above-mentioned network adaptation method between devices, on the one hand, when generating the communication address between the master device and the slave device, it is based on the universal serial bus interface through which the slave device accesses the master device. Therefore, when the slave device is not plugged or unplugged from the master device, the uniqueness of the communication address between the master device and the slave device can be ensured, thereby solving the problem in the existing technology between the master device and the slave device. There is uncertainty in the communication address between them, which leads to the problem of low communication efficiency between the master device and the slave device; on the other hand, the accessed slave can be obtained from the kernel node of the physical port associated with the master device The network node of the device; then read from the network node the interface number of the universal serial bus interface of the master device that the slave device passes through when accessing the master device; finally, based on the interface number and the current communication of the slave device Network segment, generating communication between the master device and the slave device address; therefore, when a new slave device is connected to the master device, the corresponding network node can be directly obtained and the corresponding communication address is generated based on the network node, so that the communication address between the slave device and the master device is not affected by The impact of slave device access or mounting order.

以下，将结合附图对本公开示例实施例设备间的网络适配方法进行详细的解释以及说明。Below, the network adaptation method between devices according to the exemplary embodiments of the present disclosure will be explained and described in detail with reference to the accompanying drawings.

首先，对本公开示例实施例的应用场景进行解释以及说明。具体的，本公开示例实施例所记载的设备间的网络适配方法，可以应用于终端设备，该终端设备中可以包括一个主控设备以及一个或者多个从属设备，具体的应用场景可以参考图2所示。例如，在图2中，主控设备201可以连接从属设备202、从属设备203、从属设备204、从属设备205等等；其中，所能连接的从属设备的数量可以根据主控设备所具有的USB接口来决定，接入的从属设备的数量不超过接口数量即可，具体需要接入的数量可以根据实际需要来决定，本示例对此不做特殊限制。First, the application scenarios of the exemplary embodiments of the present disclosure are explained and described. Specifically, the network adaptation method between devices recorded in the exemplary embodiments of the present disclosure can be applied to terminal devices. The terminal device can include a master control device and one or more slave devices. For specific application scenarios, refer to Figure 2 shown. For example, in Figure 2, the master device 201 can connect the slave device 202, the slave device 203, the slave device 204, the slave device 205, etc.; wherein, the number of slave devices that can be connected can be based on the USB number of the master device. It is determined by the interface. The number of connected slave devices does not exceed the number of interfaces. The specific number of connected slave devices can be determined according to actual needs. This example does not impose special restrictions on this.

但是，在图2所示出的主控设备以及从属设备中，主控设备以及从属设备中都需要设置智能***，因此主控设备以及从属设备的USB都默认工作模式均为Host模式(主机模式)；因此，若主控设备与从属设备直接通过USB进行连接，主控设备无法通过USB接口控制从属设备执行相应的指令；同时，当主控设备所连接的从属设备包括多个时，从属设备对于主控设备来说，是如何进行区分的以及如何进行通信的，是本申请主要需要解决的问题。However, in the master device and slave devices shown in Figure 2, intelligent systems need to be set up in both the master device and the slave device. Therefore, the default working mode of the USB of the master device and the slave device is Host mode (host mode). ); Therefore, if the master device and the slave device are directly connected through USB, the master device cannot control the slave device to execute the corresponding instructions through the USB interface; at the same time, when the master device is connected to multiple slave devices, the slave device For the main control device, how to distinguish and communicate are the main problems that need to be solved in this application.

为了解决上述问题，参考图3所示，在主控设备中增加了设备管理模块Device Manager301和路由管理模块Route Manager302。具体来说：首先，可以通过设备管理模块将从属设备设置为RNDIS模式。其中，RNDIS，也即Remote Network Driver Interface Specification，远程网络驱动程序接口规范；其中，RNDIS的作用为TCP(Transmission Control Protocol，传输控制协议)/IP(Internet Protocol，网际协议)over USB，也即将USB(Universal Serial Bus，通用串行总线)设备作为网卡，从而使主控设备可以通过USB设备连接网络。同时，实现USB RNDIS功能需要具备以下条件：一方面，主控设备需要支持USB Client并且USB驱动已经完成；另一方面，需要在主控设备上面安装Active Sync软件，该软件用于支持RNDIS驱动。In order to solve the above problems, as shown in Figure 3, the device management module Device Manager301 and the routing management module Route Manager302 are added to the main control device. Specifically: First, the slave device can be set to RNDIS mode through the device management module. Among them, RNDIS, also known as Remote Network Driver Interface Specification, remote network driver interface specification; among them, RNDIS functions as TCP (Transmission Control Protocol)/IP (Internet Protocol) over USB, also known as USB (Universal Serial Bus, Universal Serial Bus) device acts as a network card, so that the main control device can connect to the network through the USB device. At the same time, the following conditions are required to implement the USB RNDIS function: on the one hand, the main control device needs to support USB Client and the USB driver has been completed; on the other hand, the Active Sync software needs to be installed on the main control device, which is used to support the RNDIS driver.

在此前提下，可以通过设备管理模块将从属设备配置为RNDIS模式；在RNDIS模式下，对于从板来说，从属设备的USB口就工作在target模式；对于主控设备来说，每个从属设备就变成了一个虚拟的网卡，这样主控设备和从属设备的通讯就成为主控设备通过USB网卡和从属设备进行通信；因此，为了实现主控设备与从属设备之间的通信，需要给主控设备的网卡和从属设备的IP设置成同一网段；例如，均设置为192.168.1.0网段；同时，对于多个从属设备来说，可以分别设置成多个不同网段，具体可以参考图4所示。进一步的，设备管理模块还负责从属设备的接入或挂载。具体的，设备开机后，由于从属设备上电的顺序有先后，所以对于主控设备来说，每个从属设备挂载的顺序具有随机性；其 中，如下表1以及表2皆为可能出现的情况：Under this premise, the slave device can be configured into RNDIS mode through the device management module; in RNDIS mode, for the slave board, the USB port of the slave device works in target mode; for the master device, each slave The device becomes a virtual network card, so that the communication between the master device and the slave device becomes the communication between the master device and the slave device through the USB network card; therefore, in order to realize the communication between the master device and the slave device, it is necessary to The network card of the master device and the IP of the slave device are set to the same network segment; for example, both are set to the 192.168.1.0 network segment; at the same time, for multiple slave devices, they can be set to multiple different network segments. For details, please refer to As shown in Figure 4. Furthermore, the device management module is also responsible for accessing or mounting slave devices. Specifically, after the device is powered on, since the slave devices are powered on in sequence, for the master device, the order in which each slave device is mounted is random; , the following Table 1 and Table 2 are possible situations:

表1主板USB网卡与从板对应关系：
Table 1 Correspondence between the mainboard USB network card and the slave board:

表2主板USB网卡与从板对应关系：
Table 2 Correspondence between the mainboard USB network card and the slave board:

基于上述表1以及表2所记载的内容可以得知，由于从属设备的接入顺序的不确定性，因而主控设备与从属设备之间的通信地址的配置关系就无法确定；基于此，本公开示例实施例首先提出了一种如图1所示的设备间的网络适配方法。具体的：Based on the contents recorded in Table 1 and Table 2 above, it can be known that due to the uncertainty of the access sequence of the slave devices, the configuration relationship of the communication addresses between the master device and the slave devices cannot be determined; based on this, this paper The disclosed example embodiments first propose a network adaptation method between devices as shown in FIG. 1 . specific:

在步骤S110中，从与所述主控设备关联的物理端口的内核节点中获取接入的从属设备的网络节点。In step S110, obtain the network node of the accessed slave device from the kernel node of the physical port associated with the master device.

在本示例实施例中，在从内核节点中获取从属设备的网络节点之前，首先需要在内核节点中创建从属设备的网络节点。其中，网络节点的具体创建方法可以通过如下方式来实现：首先，启动预设程序进程，并基于所述预设程序进程对所述主控设备的物理端口的内核节点进行轮询监测；其次，在监测到所述物理端口中有从属设备接入时，在所述内核节点中创建与接入的从属设备关联的网络节点。In this example embodiment, before obtaining the network node of the slave device from the kernel node, it is first necessary to create the network node of the slave device in the kernel node. The specific creation method of the network node can be implemented in the following manner: first, start a preset program process, and perform polling monitoring of the kernel node of the physical port of the main control device based on the preset program process; secondly, When it is detected that a slave device is connected to the physical port, a network node associated with the connected slave device is created in the kernel node.

具体的，首先，可以定义一个预设进程，并在主控设备启动时，即启动该预设进程，进而通过该预设进程对主控设备的物理端口的内核节点进行轮询监测；其中，该预设进程例如可以是守护进程Deamon；当然，也可以是其他的可以用于进行物理端口以及内核节点监测的进程，本示例对此不做特殊限制。进一步的，主控设备的物理端口具体可以如下所示：Specifically, first, a preset process can be defined, and when the main control device starts, the preset process is started, and then the kernel node of the physical port of the main control device is polled and monitored through the preset process; where, The default process can be, for example, the daemon process Deamon; of course, it can also be other processes that can be used to monitor physical ports and kernel nodes. This example has no special restrictions on this. Further, the physical port of the main control device can be as follows:

物理USB1：/sys/bus/usb/device/xxxx-1.1/；Physical USB1:/sys/bus/usb/device/xxxx-1.1/;

物理USB2：/sys/bus/usb/device/xxxx-1.2/；Physical USB2:/sys/bus/usb/device/xxxx-1.2/;

进一步的，当监测到物理端口中有从属设备接入时，即可在该物理端口的内核(Kernel)节点中创建网络(net)节点；并且，当守护进程监测到内核节点中的网络节点创建成功以后，即可认为该物理端口已经有从属设备成功接入。更进一步的，当守护进程监测到有从 属设备成功接入以后，即可从对应的主控设备的物理端口下获取该从属设备的网络节点。Further, when it is detected that a slave device is connected to the physical port, a network (net) node can be created in the kernel node of the physical port; and, when the daemon process detects the creation of a network node in the kernel node, After success, it can be considered that the physical port has been successfully accessed by a slave device. Furthermore, when the daemon detects that there is a slave After the slave device is successfully connected, the network node of the slave device can be obtained from the physical port of the corresponding master device.

在步骤S120中，从所述网络节点中读取所述从属设备在接入所述主控设备时，所经由的主控设备的通用串行总线接口的接口编号。In step S120, the interface number of the universal serial bus interface of the master device through which the slave device passes when accessing the master device is read from the network node.

具体的，读取网络节点的内容，其中，网络节点的内容中包括从属设备在接入主控设备时，所经由的主控设备的通用串行总线接口的接口编号，也即主控设备为该接入的从属设备的网卡编码。其中，具体的网卡编码可以如下所示：Specifically, read the content of the network node, where the content of the network node includes the interface number of the universal serial bus interface of the master device through which the slave device accesses the master device, that is, the master device is The network card code of the accessed slave device. Among them, the specific network card encoding can be as follows:

物理USB1→网卡USB2→从卡设备192.168.1.1；Physical USB1→Network card USB2→Slave card device 192.168.1.1;

物理USB2→网卡USB3→从卡设备192.168.2.1。Physical USB2→network card USB3→slave card device 192.168.2.1.

在步骤S130中，基于所述接口编号以及所述从属设备所具有的当前通信网段，生成所述主控设备与所述从属设备之间的通信地址。In step S130, a communication address between the master device and the slave device is generated based on the interface number and the current communication network segment of the slave device.

具体的，当得到接口编号以后，即可根据该接口编号以及当前通信网段生成相应的通信地址。其中，所生成的通信地址具体可以如下所示：Specifically, after the interface number is obtained, the corresponding communication address can be generated based on the interface number and the current communication network segment. Among them, the generated communication address can be as follows:

ifconfig usb2 192.168.1.2；和/或ifconfig usb2 192.168.1.2; and/or

ifconfig usb3 192.168.2.2。ifconfig usb3 192.168.2.2.

至此可以毫无疑问的得出，本公开示例实施例所记载的设备间的网络适配方法，通过定义一个守护进程，实时去监控主板物理端口，及内核对应USB物理端口的节点，进而在确定物理端口的目录底下存在net节点时，读取net节点的内容得到主控设备的网卡编码(USB2，USB3)，进而基于网卡编码生成通信地址，使得主控设备与从属设备之间可以存在一个唯一的通信地址，且主控设备与从属设备之间的通信地址不受从属设备的挂载顺序的限制，直接基于该通信地址通信即可，进而达到提高通信效率的目的；并且，由于主控设备与从属设备之间的通信地址中包括从属设备的通信网段以及主控设备的USB接口的接口编号，进而使得主控设备可以基于该通信地址区分所接入的从属设备，解决了现有技术中主控设备无法对接入的从属设备进行区分的问题。At this point, it can be undoubtedly concluded that the network adaptation method between devices recorded in the exemplary embodiments of the present disclosure monitors the motherboard physical port and the kernel node corresponding to the USB physical port in real time by defining a daemon process, and then determines the When there is a net node under the directory of the physical port, read the content of the net node to obtain the network card code (USB2, USB3) of the master device, and then generate a communication address based on the network card code, so that there can be a unique link between the master device and the slave device. The communication address, and the communication address between the master control device and the slave device is not restricted by the mounting order of the slave device, and communication can be directly based on the communication address, thereby achieving the purpose of improving communication efficiency; and, because the master control device The communication address with the slave device includes the communication network segment of the slave device and the interface number of the USB interface of the master device, so that the master device can distinguish the connected slave devices based on the communication address, solving the problem of the existing technology The main control device cannot distinguish the connected slave devices.

进一步的，为了避免主控设备中由于接入过多从属设备进而导致的主控设备负载过重的问题，该设备间的网络适配方法还包括：首先，在所述主控设备中建立与接入的从属设备关联的虚拟节点，并计算已经建立的虚拟节点的节点数量；其次，判断所述节点数量是否大于所述主控设备所具有的通用串行总线接口的接口数量；然后，在确定所述节点数量大于等于所述接口数量时，终止所述从属设备的接入。也即，在具体的应用过程中，在监测到存在从属设备接入时，即可在主控设备对应的物理端口的内核节点中创建一个与该接入的从属设备对应的虚拟节点；其中，创建一个虚拟节点就表示已经有一个从属设备接入，间隔一定的时间计算虚拟节点的节点数量与主控设备所具有的通用串行总线接口的接口数量之间的差值，若节点数量已经达到接口数量，则需要终止从属设备的接入。此处需要补充说明的是，此处之所以限定从属设备的接入数量，是为了确保从属设备与主控设备之间的通信地址的唯一性(一个通信地址中包括一个唯一确定的接口编码，若一个接口中接入多个从属设备，则无法保证通信地址的唯一性，进而无法达到提高通信效率的目的)； 也即，本公开示例实施例所记载的设备间的网络适配方法，丢弃USB网卡编号与从板的对应关系，而改为设备实际物理接口与从板的对应关系，进而实现相应的目的。Further, in order to avoid the problem of overloading the main control device due to the access of too many slave devices in the main control device, the network adaptation method between the devices also includes: first, establishing a connection with the main control device in the main control device. The virtual node associated with the accessed slave device, and calculate the number of nodes of the established virtual node; secondly, determine whether the number of nodes is greater than the number of universal serial bus interfaces of the master control device; then, When it is determined that the number of nodes is greater than or equal to the number of interfaces, the access of the slave device is terminated. That is to say, during the specific application process, when the access of a slave device is detected, a virtual node corresponding to the accessed slave device can be created in the kernel node of the physical port corresponding to the master device; where, Creating a virtual node means that there is already a slave device connected. Calculate the difference between the number of nodes of the virtual node and the number of universal serial bus interfaces of the master device at a certain interval. If the number of nodes has reached If the number of interfaces is too high, the access of the slave device needs to be terminated. What needs to be added here is that the reason why the number of access slave devices is limited here is to ensure the uniqueness of the communication address between the slave device and the master device (a communication address includes a unique interface code, If multiple slave devices are connected to an interface, the uniqueness of the communication address cannot be guaranteed, and the purpose of improving communication efficiency cannot be achieved); That is to say, the network adaptation method between devices described in the exemplary embodiments of the present disclosure discards the corresponding relationship between the USB network card number and the slave board, and replaces it with the corresponding relationship between the actual physical interface of the device and the slave board, thereby achieving the corresponding purpose.

更进一步的，当得到通信地址以后，即可基于该通信地址建立主控设备与从属设备之间的通信。具体的，可以通过如下方式实现：根据所述通信地址建立所述主控设备与从属设备之间的通信连接，并基于所述通信连接，在所述主控设备以及从属设备之间进行数据传输和/或指令传输。其中，在建立主控设备以及从属设备之间的通信连接以后，主控设备即可控制从属设备执行相应的任务和/或传输相应的数据；并且，在不同的应用场景下，所传输的数据和/或所执行的指令不同，本申请对此不做特殊限定。Furthermore, after obtaining the communication address, communication between the master device and the slave device can be established based on the communication address. Specifically, this can be achieved in the following manner: establishing a communication connection between the master device and the slave device according to the communication address, and performing data transmission between the master device and the slave device based on the communication connection. and/or instruction transmission. Among them, after the communication connection between the master device and the slave device is established, the master device can control the slave device to perform corresponding tasks and/or transmit corresponding data; and, in different application scenarios, the transmitted data and/or the instructions executed are different, this application does not make special limitations on this.

在一种示例实施例中，在为每个从属设备分配完主控设备与从属设备之间的通信地址后，一个新的问题就会出现：也即，由于主控设备上还有连接外网的网卡，比如Ethernet(以太网)和/或4G网卡等，而且4G网卡通常也已USB0-X这种方式命名；但是，从卡连接后，对于主控设备来说，又多出了许多网卡，这样对于需要走外网的数据包，主控设备就不知道该控制哪个网卡给转发出去，一旦选择错了网卡，直接的后果就是设备不能上网。基于此，本公开示例实施例还提供了另一种设备间的网络适配方法。具体的，参考图5所示，该设备间的网络适配方法还可以包括以下步骤：In an example embodiment, after each slave device is assigned a communication address between the master device and the slave device, a new problem will arise: that is, because the master device is still connected to the external network Network cards, such as Ethernet and/or 4G network cards, etc., and 4G network cards are usually named in this way USB0-X; however, after the slave card is connected, there are many more network cards for the master device , so that for data packets that need to go to the external network, the main control device does not know which network card to control and forward them. Once the wrong network card is selected, the direct consequence is that the device cannot access the Internet. Based on this, example embodiments of the present disclosure also provide another network adaptation method between devices. Specifically, as shown in Figure 5, the network adaptation method between devices may also include the following steps:

步骤S510，根据所述通信地址，建立所述主控设备的通用串行总线接口与所述从属设备之间的当前设备映射关系表；Step S510: Establish a current device mapping relationship table between the universal serial bus interface of the master device and the slave device according to the communication address;

步骤S520，判断所述当前设备映射关系表与历史设备映射关系表之间是否存在新接入的从属设备；Step S520, determine whether there is a newly accessed slave device between the current device mapping relationship table and the historical device mapping relationship table;

步骤S530，若存在新接入的从属设备，则为新接入的从属设备增加路由路径；Step S530: If there is a newly accessed slave device, add a routing path for the newly accessed slave device;

步骤S540，若不存在新接入的从属设备，则对所述历史设备映射关系表中所包括的所有路由路径进行刷新。Step S540: If there is no newly accessed slave device, refresh all routing paths included in the historical device mapping relationship table.

以下，将对步骤S510-步骤S540进行解释以及说明。具体的，由于主控设备中还包括了路由管理模块；因此，为了可以确保主控设备所选择的网卡可以上网，需要为新增的网卡增加路由路径，或者对设备映射关系表中所包括的所有路由路径进行刷新，以确保不管主控设备选择了哪一条路由路径，均可以实现上网的目的。进一步的，当路由路径增加完成或者刷新完成后，该设备间的网络适配方法还包括：根据所述路由路径建立所述主控设备和/或从属设备与外部应用程序之间的通信连接。同时，在具体的应用过程中，路由管理模块可以通过对多个从卡设置路由来实现设备的上网。它的具体过程是：启动路由管理的守护进程和定时器；其中，路由管理的守护进程与上述网络节点监测的守护进程可以是同一个，也可以是独立设置的，本示例对此不做特殊限制；也即，启动守护进程以后，通过定时器记录间隔时长，若时长达到规定时长则查看DeviceManager的设备映射表(当前设备映射关系表)；然后，对于新增的网卡(从属设备)增加一条路由路径；若没有增加，为所有的设备刷新路由路径表。通过该方法，可以避免由于挂载了较多网卡(从属设备和/或Ethernet(以太网)和/或4G网卡)，且给每个网卡设置了不同网段的IP，进而导致主 控设备中存在多个IP，若未给从属设备配置路由路径，会导致主控设备通过该从属设备对应的网卡向外发送上网请求，进而由于该从属设备的网卡不存在路由路径会使得上网失败的问题。In the following, steps S510 to S540 will be explained and described. Specifically, since the main control device also includes a routing management module; therefore, in order to ensure that the network card selected by the main control device can access the Internet, it is necessary to add a routing path for the new network card, or to configure the routing path included in the device mapping relationship table. All routing paths are refreshed to ensure that no matter which routing path is selected by the main control device, the purpose of accessing the Internet can be achieved. Further, when the routing path is added or refreshed, the network adaptation method between devices further includes: establishing a communication connection between the master device and/or the slave device and the external application according to the routing path. At the same time, in the specific application process, the routing management module can realize the device's Internet access by setting routes for multiple slave cards. The specific process is: start the daemon process and timer of routing management; among them, the daemon process of routing management and the daemon process of the above-mentioned network node monitoring can be the same, or they can be set independently. This example does not make any special arrangements for this. Limit; that is, after starting the daemon process, record the interval length through a timer. If the time reaches the specified length, check the device mapping table of DeviceManager (current device mapping table); then, add a new network card (slave device) for the new network card Routing path; if not added, refresh the routing path table for all devices. Through this method, you can avoid mounting more network cards (slave devices and/or Ethernet (Ethernet) and/or 4G network cards), and setting IPs of different network segments for each network card, which will cause the master to There are multiple IPs in the control device. If the routing path is not configured for the slave device, the master device will send an Internet access request through the network card corresponding to the slave device. In addition, because the network card of the slave device does not have a routing path, Internet access will fail. The problem.

以下，结合具体的应用场景对本公开示例实施例的设备间的网络适配方法进行解释以及说明。具体的，以终端设备为智能冰箱为例，对本公开示例实施例的具体实现过程进行解释以及说明。其中，此处所记载的智能冰箱，可以用于指代具有AI(Artificial Intelligence)功能的智能冰箱。具体的，参考图6所示，冰箱门为一个显示界面601，用于显示冰箱内的商品信息，另外冰箱上连接有多个摄像头，冰箱外面所设置的智能摄像头602，可以采集冰箱前的人的特征信息，包括年龄、性别等，冰箱里面所设置的智能摄像头603，可以实时监测放入商品的信息，和冰箱内的所有商品信息。进一步的，智能冰箱上还设置有一个主控SoC(System on a Chip，***级芯片)板卡(主控设备)，该主控SoC板卡可以被认为是智能冰箱的主控SoC板卡，运行有Android***，可以驱动液晶、智能摄像头和其他外设；同时，每个智能摄像头(从属设备)由一个小型的SoC板卡进行驱动，从属设备上运行有小型的Linux***，在上面可以部署人脸识别、商品识别的算法。本发明为了解决，这种主、从SoC板卡的部署及通讯问题。In the following, the network adaptation method between devices according to the exemplary embodiments of the present disclosure will be explained and described in conjunction with specific application scenarios. Specifically, taking the terminal device as a smart refrigerator as an example, the specific implementation process of the exemplary embodiments of the present disclosure will be explained and described. Among them, the smart refrigerator described here can be used to refer to a smart refrigerator with AI (Artificial Intelligence) function. Specifically, as shown in Figure 6, the refrigerator door is a display interface 601 for displaying product information in the refrigerator. In addition, multiple cameras are connected to the refrigerator. A smart camera 602 installed outside the refrigerator can collect information about people in front of the refrigerator. Characteristic information, including age, gender, etc., the smart camera 603 installed in the refrigerator can monitor the information of the products placed in the refrigerator in real time, as well as the information of all products in the refrigerator. Furthermore, the smart refrigerator is also equipped with a main control SoC (System on a Chip, system-level chip) board (main control device). This main control SoC board can be considered as the main control SoC board of the smart refrigerator. It runs an Android system and can drive LCD, smart cameras and other peripherals; at the same time, each smart camera (slave device) is driven by a small SoC board, and the slave device runs a small Linux system on which it can be deployed Algorithms for face recognition and product recognition. The present invention aims to solve the deployment and communication problems of such master and slave SoC boards.

进一步的，在智能冰箱的场景下，基于所述通信连接，在所述主控设备以及从属设备之间进行数据传输和/或指令传输，具体可以通过如下方式实现：首先，根据所述从属设备在所述终端设备中所处的当前位置，在所述从属设备中部署计算机视觉算法；其次，控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，并基于所述通信连接图像识别结果从从属设备传输至主控设备；其中，所述计算机视觉算法包括人像识别算法和/或物品识别算法；所述当前位置包括智能冰箱内部和/或智能冰箱外部。Further, in the scenario of a smart refrigerator, based on the communication connection, data transmission and/or instruction transmission is performed between the master control device and the slave device, which can be implemented in the following manner: First, according to the slave device Deploy a computer vision algorithm in the slave device at the current position of the terminal device; secondly, control the slave device to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result, And based on the communication connection, the image recognition result is transmitted from the slave device to the master device; wherein the computer vision algorithm includes a portrait recognition algorithm and/or an object recognition algorithm; the current location includes the inside of the smart refrigerator and/or the outside of the smart refrigerator. .

在一种示例实施例中，根据所述从属设备在所述终端设备中所处的当前位置，在所述从属设备中部署计算机视觉算法，具体可以通过如下方式实现：当所述智能摄像头所处的当前位置位于智能冰箱内部时，在该智能摄像头中部署物品识别算法；当所述智能摄像头所处的当前位置位于智能冰箱内部时，在该智能摄像头中部署人像识别算法。In an example embodiment, deploying a computer vision algorithm in the slave device according to the current location of the slave device in the terminal device can be implemented in the following manner: when the smart camera is in When the current location of the smart camera is inside the smart refrigerator, an object recognition algorithm is deployed in the smart camera; when the current location of the smart camera is inside the smart refrigerator, a portrait recognition algorithm is deployed in the smart camera.

在一种示例实施例中，控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，包括：主控设备在检测到与所述智能冰箱对应的第一预设位置存在目标对象时，根据主控设备与从属设备之间的当前设备映射关系表，获取处于该第一预设位置的从属设备与主控设备之间的通信地址；主控设备生成与处于第一预设位置的从属设备关联的人脸图像采集指令，并基于处于第一预设位置的从属设备与主控设备之间的通信地址，将所述人脸图像采集指令发送至处于第一预设位置的从属设备；处于第一预设位置的从属设备响应于所述人脸图像采集指令，采集所述目标对象的待识别人脸图像，并利用预设的人像识别算法对采集到的待识别人脸图像进行识别，得到人像识别结果。In an example embodiment, controlling the slave device to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result includes: the master control device detects the first image corresponding to the smart refrigerator. When the target object exists at the preset position, the communication address between the slave device and the master device at the first preset position is obtained according to the current device mapping relationship table between the master device and the slave device; the master device generates and Face image collection instructions associated with the slave device at the first preset position, and based on the communication address between the slave device at the first preset position and the master control device, send the face image collection instructions to the slave device at the first preset position. A slave device at a preset position; the slave device at the first preset position responds to the face image collection instruction, collects the face image to be recognized of the target object, and uses a preset face recognition algorithm to collect the collected face image. Recognize the face image to be recognized and obtain the portrait recognition result.

其中，利用预设的人像识别算法对采集到的待识别人脸图像进行识别，得到人像识别 结果，包括：首先，利用预设的人脸检测与关键点定位工具，检测所述待识别人脸图像的待识别人脸区域；其次，在所述待识别人脸区域中提取所述待识别人脸图像的待识别脸部关键点，并根据所述待识别脸部关键点，计算所述待识别人脸特征；然后，计算所述待识别人脸特征以及处于第二预设位置处的从属设备中的特征值库中的原始人脸特征之间的欧式距离；进一步的，根据所述欧式距离，得到所述待识别人脸特征与所述原始人脸特征之间的相似度比对结果，并根据所述相似度比对结果得到所述人像识别结果。Among them, the preset portrait recognition algorithm is used to recognize the collected face images to be recognized, and the portrait recognition is obtained. The results include: first, using preset face detection and key point positioning tools to detect the face area to be recognized in the face image to be recognized; secondly, extracting the face area to be recognized in the face area to be recognized Key points of the face to be recognized in the face image, and based on the key points of the face to be recognized, calculate the facial features to be recognized; then, calculate the facial features to be recognized and the facial features at the second preset position The Euclidean distance between the original facial features in the feature value library in the slave device; further, based on the Euclidean distance, a similarity comparison between the facial features to be recognized and the original facial features is obtained As a result, the portrait recognition result is obtained according to the similarity comparison result.

以下，将对人脸识别的具体实现过程进行解释以及说明。具体的，人脸识别(Facial Recognition)，就是通过设置在智能冰箱的外部的智能摄像头获取用户的待识别人脸图像，再利用设置在该智能摄像头中的人脸识别算法对其脸部的五官位置、脸型和角度进行计算分析，进而和自身数据库里已有的范本进行比对，后判断出用户的真实身份。进一步的，在具体的人脸识别过程中，首先，需要对局部区域进行定义；其次，人脸局部区域特征的提取，依据经过样本训练后得到的变换矩阵将人脸图像向量映射为人脸特征向量；进一步的，局部特征选择(可选)；最后一步是进行分类。分类器多采用组合分类器的形式，每个局部特征对应一个分类器，后可用投票或线性加权等方式得到终识别结果。Below, the specific implementation process of face recognition will be explained and described. Specifically, facial recognition (Facial Recognition) is to obtain the user's face image to be recognized through a smart camera set outside the smart refrigerator, and then use the face recognition algorithm set in the smart camera to identify the facial features of the user. The position, face shape and angle are calculated and analyzed, and then compared with existing templates in its own database to determine the user's true identity. Furthermore, in the specific face recognition process, firstly, the local area needs to be defined; secondly, to extract the features of the local area of the face, the face image vector is mapped into the face feature vector based on the transformation matrix obtained after sample training. ;Further, local feature selection (optional); The last step is classification. Classifiers mostly take the form of combined classifiers. Each local feature corresponds to a classifier, and the final recognition result can be obtained by voting or linear weighting.

在一种示例实施例中，控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，还可以通过如下方式实现：首先，主控设备在检测到与所述智能冰箱对应的第二预设位置存在新增物品时，根据主控设备与从属设备之间的当前设备映射关系表，获取处于该第二预设位置的从属设备与主控设备之间的通信地址；其次，主控设备生成与处于第二预设位置的从属设备关联的物品图像采集指令，并基于处于第二预设位置的从属设备与主控设备之间的通信地址，将所述物品图像采集指令发送至处于第二预设位置的从属设备；最后，处于第二预设位置的从属设备响应于所述物品图像采集指令，采集所述新增物品的待识别物品图像，并利用预设的物品识别算法对采集到的待识别物品图像进行识别，得到物品图像识别结果。其中，在对物品进行识别的过程中所采用倒的物品识别算法可以包括卷积神经网络模型、循环神经网络模型、深度神经网络模型以及YOLO系列模型，例如YOLOV2、YOLOV3、YOLOV4以及YOLOV5等等，本示例对此不做特殊限制。In an example embodiment, controlling the slave device to recognize the collected image to be recognized based on the computer vision algorithm to obtain the image recognition result can also be achieved in the following manner: first, the master device detects the When there is a new item in the second preset position corresponding to the smart refrigerator, according to the current device mapping relationship table between the master device and the slave device, the relationship between the slave device and the master device at the second preset position is obtained. communication address; secondly, the master control device generates an item image collection instruction associated with the slave device at the second preset position, and based on the communication address between the slave device at the second preset position and the master control device, the master control device The item image collection instruction is sent to the slave device at the second preset position; finally, the slave device at the second preset position responds to the item image collection instruction, collects the to-be-identified item image of the newly added item, and uses The preset item recognition algorithm recognizes the collected image of the item to be identified and obtains the item image recognition result. Among them, the object recognition algorithms used in the process of identifying objects can include convolutional neural network models, recurrent neural network models, deep neural network models and YOLO series models, such as YOLOV2, YOLOV3, YOLOV4 and YOLOV5, etc. This example imposes no special restrictions on this.

以下，以物品识别算法为为例，对具体的识别过程进行解释以及说明。具体的，参考图7所示，该YOLOV4模型可以包括输入层701、主干特征提取网络(BackBone)702、颈部特征融合网络(Neck)703以及头部特征检测网络(Head)704以及输出层705。其中，输入层701、主干(Back Bone)特征提取网络702、颈部(Neck)特征融合网络703以及头部(Head)特征检测网络704以及输出层705依次连接。Below, the specific identification process is explained and illustrated, taking the item recognition algorithm as an example. Specifically, as shown in Figure 7, the YOLOV4 model may include an input layer 701, a backbone feature extraction network (BackBone) 702, a neck feature fusion network (Neck) 703, a head feature detection network (Head) 704, and an output layer 705. . Among them, the input layer 701, the back bone feature extraction network 702, the neck feature fusion network 703, the head feature detection network 704 and the output layer 705 are connected in sequence.

其中，参考图8所示，主干特征提取网络中包括CBM模块801以及多个CSP模块，所述多个CSP模块包括第一个CSP模块802、第二个CSP模块803、第三个CSP模块804、第四个CSP模块805以及第五个CSP模块806。也即，本公开示例实施例所采用的Backbone，可以基于CSPDarknet53实现，该CSPDarknet53中包含了1个CBM模块和5个CSP模块； 其中，CBM模块可以由Conv+Bn+Mish激活函数三者组成；Conv为Convolutional卷积，Bn为Batch Normalization；Mish为激活函数；CSP模块可以由CBM模块以及一个或者多个Res Unit模块concate组成。其中，CSP可以用于表示Cross Stage Partial，也即其是一个增强学习能力的跨阶段局部网络。Among them, as shown in Figure 8, the backbone feature extraction network includes a CBM module 801 and multiple CSP modules. The multiple CSP modules include a first CSP module 802, a second CSP module 803, and a third CSP module 804. , the fourth CSP module 805 and the fifth CSP module 806. That is to say, the Backbone used in the exemplary embodiment of the present disclosure can be implemented based on CSPDarknet53, which includes 1 CBM module and 5 CSP modules; Among them, the CBM module can be composed of Conv+Bn+Mish activation function; Conv is Convolutional convolution, Bn is Batch Normalization; Mish is the activation function; CSP module can be composed of CBM module and one or more Res Unit modules concate . Among them, CSP can be used to represent Cross Stage Partial, that is, it is a cross-stage partial network that enhances learning capabilities.

在一种示例实施例中，第一个CSP模块802可以由CBM模块和1个Resunint模块Concate组成。在具体的应用过程中，可以将CBM模块处理后的特征图608*608*32的F_conv2传入第一个CSP模块进行处理(其中只有1个残差单元)。In an example embodiment, the first CSP module 802 may be composed of a CBM module and a Resunint module Concate. In the specific application process, the F_conv2 of the 608*608*32 feature map processed by the CBM module can be passed into the first CSP module for processing (there is only one residual unit).

在一种示例实施例中，第二个CSP模块803可以由CBM模块和2个Resunint模块Concate组成。在具体的应用过程中，可以将第一个CSP模块处理后的304*304*64特征图，传入到第二个CSP模块处理。具体的，经过下采样后变成了152*152*128的特征图，然后分别经过两个1*1*64的s＝1的卷积后得到两个分支，其中一个分支的特征块进入残差模块进行处理后，再与另一个分支进行拼接；其中，第二个CSP模块的最后输出是152*152*128(残差模块中的卷积层：1*1*64和3*3*64)。In an example embodiment, the second CSP module 803 may be composed of a CBM module and two Resunint modules Concate. In the specific application process, the 304*304*64 feature map processed by the first CSP module can be passed to the second CSP module for processing. Specifically, after downsampling, it becomes a 152*152*128 feature map, and then two branches are obtained after two 1*1*64 s=1 convolutions, and the feature block of one branch enters the residual After processing by the difference module, it is spliced with another branch; among them, the final output of the second CSP module is 152*152*128 (the convolutional layer in the residual module: 1*1*64 and 3*3* 64).

在一种示例实施例中，第三个CSP模块804可以由8个Res Unint模块和CBM模块Concate组成。在具体的应用过程中，可以将第二个CSP模块处理后的152*152*128的特征图，传入到第三个CSP模块处理，最终第三个csp模块的最后输出是76*76*256(残差模块中的卷积层：1*1*128和3*3*128)，这个模块后又分为两个分支，一个分支继续进行第四个CSP模块处理，另一个分支直接进入到Neck处理。In an example embodiment, the third CSP module 804 may be composed of 8 Res Unint modules and a CBM module Concate. In the specific application process, the 152*152*128 feature map processed by the second CSP module can be passed to the third CSP module for processing, and the final output of the third CSP module is 76*76* 256 (convolutional layer in the residual module: 1*1*128 and 3*3*128), this module is divided into two branches, one branch continues to process the fourth CSP module, and the other branch directly enters Go to Neck for processing.

在一种示例实施例中，第四个CSP模块805由8个Res Unint模块和CBM模块Concate组成。在具体的应用过程中，可以将第三个CSP模块的一个分支76*76*256的特征图，传入到第四个CSP模块处理，最终第四个CSP模块的最后输出是38*38*512(残差模块中的卷积层：1*1*256和3*3*256)，这个模块后又分为两个分支，一个分支继续进行第五个CSP模块处理，另一个分支直接进入到Neck处理。In an example embodiment, the fourth CSP module 805 is composed of 8 Res Unint modules and the CBM module Concate. In the specific application process, the 76*76*256 feature map of a branch of the third CSP module can be passed to the fourth CSP module for processing, and the final output of the fourth CSP module is 38*38* 512 (convolutional layer in the residual module: 1*1*256 and 3*3*256), this module is divided into two branches, one branch continues to process the fifth CSP module, and the other branch directly enters Go to Neck for processing.

在一种示例实施例中，第五个CSP模块806由4个Res Unint模块和CBM模块Concate组成。在具体的应用过程中，可以将第四个CSP模块的一个分支38*38*512的特征图，传入到第五个CSP模块处理，最终第五个csp模块的最后输出是19*19*1024(残差模块中的卷积层：1*1*512和3*3*512)，这个模块输出结果直接进入到Neck处理。In an example embodiment, the fifth CSP module 806 consists of 4 Res Unint modules and a CBM module Concate. In the specific application process, the 38*38*512 feature map of a branch of the fourth CSP module can be passed to the fifth CSP module for processing, and the final output of the fifth CSP module is 19*19* 1024 (convolutional layer in the residual module: 1*1*512 and 3*3*512), the output result of this module directly enters Neck processing.

进一步的，参考图9所示，颈部特征融合网络(Neck)可以包括SPP模块901、多个CBL模块、多个上采样模块以及多个拼接模块；其中，所述多个CBL模块包括第一个CBL模块902、第二个CBL模块903、第三个CBL模块904、第四个CBL模块905、第五个CBL模块906、第六个CBL模块907、第七个CBL模块908、第八个CBL模块909、第九个CBL模块910、第十个CBL模块911、第十一个CBL模块912以及第十二个CBL模块913，所述多个上采样模块包括第一个上采样模块914以及第二个上采样模块915，所述多个拼接模块包括第一个拼接模块916、第二个拼接模块917、第三个拼接模块918以及第四个拼接模块919。 Further, as shown in Figure 9, the neck feature fusion network (Neck) may include an SPP module 901, multiple CBL modules, multiple upsampling modules and multiple splicing modules; wherein the multiple CBL modules include a first The first CBL module 902, the second CBL module 903, the third CBL module 904, the fourth CBL module 905, the fifth CBL module 906, the sixth CBL module 907, the seventh CBL module 908, the eighth CBL module 909, ninth CBL module 910, tenth CBL module 911, eleventh CBL module 912 and twelfth CBL module 913. The plurality of upsampling modules include a first upsampling module 914 and The second upsampling module 915, the plurality of splicing modules include a first splicing module 916, a second splicing module 917, a third splicing module 918 and a fourth splicing module 919.

其中，CBL模块可以由Conv+Bn+Leaky_relu激活函数三者组成；SPP(Spatial Pyramid Pooling，空间金字塔池化结构)模块的前后三个CBL模块(第一个CBL模块以及第二个CBL模块，第一个CBL模块以及第二个CBL模块中分别包括3个)是对称的，它们的卷积分别是1*1*512,3*3*1024和1*1*512，步长都是1；SPP模块可以采用1×1，5×5，9×9，13×13的最大池化的方式，进行多尺度融合；其中，SPP模块采用的1×1，5×5padding＝5//2，9×9padding＝9//2，13×13padding＝13//2的最大池化的方式，进行多尺度融合，从前面三个CBL模块输出的结果：19*19*512的特征图，将之送入SPP模块中，最后的结果为19*19*2048，再经过三个CBL模块的卷积后得到19*19*512的特征图。Among them, the CBL module can be composed of Conv+Bn+Leaky_relu activation function; the three CBL modules before and after the SPP (Spatial Pyramid Pooling, spatial pyramid pooling structure) module (the first CBL module and the second CBL module, The first CBL module and the second CBL module respectively include 3) are symmetrical, their convolutions are 1*1*512, 3*3*1024 and 1*1*512 respectively, and the step size is 1 ;The SPP module can use the maximum pooling method of 1×1, 5×5, 9×9, and 13×13 for multi-scale fusion; among them, the SPP module uses 1×1, 5×5 padding=5//2 , 9×9padding=9//2, 13×13padding=13//2 maximum pooling method, multi-scale fusion, the output results from the first three CBL modules: 19*19*512 feature map, will After being sent to the SPP module, the final result is 19*19*2048. After convolution of three CBL modules, a feature map of 19*19*512 is obtained.

更进一步的，参考图10所示，头部特征检测网络(Head)可以包括多个CBL模块以及多个卷积模块；其中，多个CBL模块可以包括第十三个CBL模块1001、第十四个CBL模块1002以及第十五个CBL模块1003；多个卷积模块可以包括第一个卷积模块1004、第二个卷积模块1005以及第三个卷积模块1006。其中，在具体的应用过程中，Yolo Head利用获得到的特征进行预测，是一个解码的过程。具体的，在特征利用部分，YoloV4提取多尺度特征进行目标检测，一共提取三个特征层，分别位于中间层，中下层，底层，三个特征层的Shape分别为(19,19,255)，(38,38,255)，(76,76,255)。三张特征图就是整个Yolo输出的检测结果，检测框位置(4维)、检测置信度(1维)、类别(80维)都在其中，加起来正好是85维。特征图最后的维度85，代表的就是这些信息，而特征图其他维度N×N×3，N×N代表了检测框的参考位置信息，3是3个不同尺度的先验框。Furthermore, as shown in Figure 10, the head feature detection network (Head) may include multiple CBL modules and multiple convolution modules; wherein the multiple CBL modules may include a thirteenth CBL module 1001, a fourteenth CBL module A CBL module 1002 and a fifteenth CBL module 1003; the plurality of convolution modules may include a first convolution module 1004, a second convolution module 1005 and a third convolution module 1006. Among them, in the specific application process, Yolo Head uses the obtained features to make predictions, which is a decoding process. Specifically, in the feature utilization part, YoloV4 extracts multi-scale features for target detection. A total of three feature layers are extracted, which are located in the middle layer, middle and lower layers, and bottom layer. The Shapes of the three feature layers are (19, 19, 255), (38). ,38,255), (76,76,255). The three feature maps are the detection results output by Yolo. The detection frame position (4 dimensions), detection confidence (1 dimension), and category (80 dimensions) are all included in them. The total is exactly 85 dimensions. The last dimension of the feature map, 85, represents this information, while the other dimensions of the feature map are N×N×3. N×N represents the reference position information of the detection frame, and 3 is the a priori frame of 3 different scales.

在具体的进行物品识别的过程中，首先，可以利用所述主干特征提取网络对所述待识别物品图像进行下采样处理，得到第一局部特征；然后，利用所述颈部特征融合网络对所述第一局部特征进行从深层到浅层、再从浅层到深层的双向融合，得到第一全局特征；最后，利用所述头部特征检测网络对所述第一全局特征中包括的目标对象的类别信息以及位置信息进行检测，得到所述图像特征。其中，各模块在图像物品识别过程中的具体处理过程，此处不再赘述。In the specific process of object recognition, first, the backbone feature extraction network can be used to downsample the image of the object to be identified to obtain the first local feature; then, the neck feature fusion network can be used to downsample the image of the object to be identified. The first local features are bidirectionally fused from deep to shallow and then from shallow to deep to obtain the first global feature; finally, the head feature detection network is used to detect the target object included in the first global feature. The category information and location information are detected to obtain the image features. Among them, the specific processing process of each module in the image object recognition process will not be described again here.

本公开示例实施例还提供了一种设备间的网络适配装置，配置于终端设备，所述终端设备中包括主控设备以及从属设。具体的，参考图11所示，该设备间的网络适配装置可以包括网络节点获取模块1110、接口编号读取模块1120以及通信地址生成模块1130。其中：Example embodiments of the present disclosure also provide a network adaptation device between devices, which is configured in a terminal device. The terminal device includes a master control device and a slave device. Specifically, referring to FIG. 11 , the network adaptation device between devices may include a network node acquisition module 1110 , an interface number reading module 1120 , and a communication address generation module 1130 . in:

网络节点获取模块1110，可以用于从与所述主控设备关联的物理端口的内核节点中获取接入的从属设备的网络节点；The network node acquisition module 1110 may be used to acquire the network node of the accessed slave device from the kernel node of the physical port associated with the master control device;

接口编号读取模块1120，可以用于从所述网络节点中读取所述从属设备在接入所述主控设备时，所经由的主控设备的通用串行总线接口的接口编号；The interface number reading module 1120 can be used to read from the network node the interface number of the universal serial bus interface of the master device through which the slave device accesses the master device;

通信地址生成模块1130，可以用于基于所述接口编号以及所述从属设备所具有的当前通信网段，生成所述主控设备与所述从属设备之间的通信地址。The communication address generation module 1130 may be configured to generate a communication address between the master device and the slave device based on the interface number and the current communication network segment of the slave device.

在本公开的一种示例性实施例中，所述设备间的网络适配方法还包括： In an exemplary embodiment of the present disclosure, the network adaptation method between devices further includes:

轮询监测模块，可以用于启动预设程序进程，并基于所述预设程序进程对所述主控设备的物理端口的内核节点进行轮询监测；The polling monitoring module can be used to start a preset program process, and perform polling monitoring of the kernel node of the physical port of the main control device based on the preset program process;

网络节点创建模块，可以用于在监测到所述物理端口中有从属设备接入时，在所述内核节点中创建与接入的从属设备关联的网络节点。The network node creation module may be configured to create, in the kernel node, a network node associated with the accessed slave device when it is detected that a slave device is connected to the physical port.

在本公开的一种示例性实施例中，所述设备间的网络适配装置还包括：In an exemplary embodiment of the present disclosure, the network adaptation device between devices further includes:

节点数量计算模块，可以用于在所述主控设备中建立与接入的从属设备关联的虚拟节点，并计算已经建立的虚拟节点的节点数量；The node number calculation module can be used to establish a virtual node associated with the accessed slave device in the main control device, and calculate the number of nodes of the established virtual node;

节点数量判断模块，可以用于判断所述节点数量是否大于所述主控设备所具有的通用串行总线接口的接口数量；A node number judgment module can be used to judge whether the number of nodes is greater than the number of interfaces of the universal serial bus interface of the main control device;

终止接入模块，可以用于在确定所述节点数量大于等于所述接口数量时，终止所述从属设备的接入。The access termination module may be configured to terminate the access of the slave device when it is determined that the number of nodes is greater than or equal to the number of interfaces.

在本公开的一种示例性实施例中，所述设备间的网络适配装置还包括：In an exemplary embodiment of the present disclosure, the network adaptation device between devices further includes:

数据传输模块，可以用于根据所述通信地址建立所述主控设备与从属设备之间的通信连接，并基于所述通信连接，在所述主控设备以及从属设备之间进行数据传输和/或指令传输。A data transmission module may be used to establish a communication connection between the master device and the slave device according to the communication address, and perform data transmission and/or between the master device and the slave device based on the communication connection. or instruction transmission.

在本公开的一种示例性实施例中，所述设备间的网络适配装置还包括：In an exemplary embodiment of the present disclosure, the network adaptation device between devices further includes:

设备映射关系表建立模块，可以用于根据所述通信地址，建立所述主控设备的通用串行总线接口与所述从属设备之间的当前设备映射关系表；A device mapping relationship table establishment module, which can be used to establish a current device mapping relationship table between the universal serial bus interface of the master control device and the slave device according to the communication address;

新接入的从属设备片段模块，可以用于判断所述当前设备映射关系表与历史设备映射关系表之间是否存在新接入的从属设备；The newly accessed slave device fragment module can be used to determine whether there is a newly accessed slave device between the current device mapping relationship table and the historical device mapping relationship table;

路由路径增加模块，可以用于若存在新接入的从属设备，则为新接入的从属设备增加路由路径，若不存在新接入的从属设备，则对所述历史设备映射关系表中所包括的所有路由路径进行刷新。The routing path adding module can be used to add a routing path for the newly accessed slave device if there is a newly accessed slave device; if there is no newly accessed slave device, add a routing path to the historical device mapping relationship table. All routing paths included are refreshed.

在本公开的一种示例性实施例中，所述设备间的网络适配装置还包括：In an exemplary embodiment of the present disclosure, the network adaptation device between devices further includes:

通信连接建立模块，可以用于根据所述路由路径建立所述主控设备和/或从属设备与外部应用程序之间的通信连接。A communication connection establishing module may be used to establish a communication connection between the master device and/or the slave device and an external application according to the routing path.

在本公开的一种示例性实施例中，所述终端设备包括智能冰箱，所述主控设备包括主控SoC板卡，所述从属设备包括一个或者多个智能摄像头；In an exemplary embodiment of the present disclosure, the terminal device includes a smart refrigerator, the main control device includes a main control SoC board, and the slave device includes one or more smart cameras;

其中，基于所述通信连接，在所述主控设备以及从属设备之间进行数据传输和/或指令传输，包括：Wherein, based on the communication connection, data transmission and/or instruction transmission is performed between the master control device and the slave device, including:

根据所述从属设备在所述终端设备中所处的当前位置，在所述从属设备中部署计算机视觉算法；Deploying a computer vision algorithm in the slave device according to the current location of the slave device in the terminal device;

控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，并基于所述通信连接图像识别结果从从属设备传输至主控设备。The slave device is controlled to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result, and the image recognition result is transmitted from the slave device to the master control device based on the communication connection.

在本公开的一种示例性实施例中，所述计算机视觉算法包括人像识别算法和/或物品 识别算法；所述当前位置包括智能冰箱内部和/或智能冰箱外部；In an exemplary embodiment of the present disclosure, the computer vision algorithm includes a portrait recognition algorithm and/or an item Recognition algorithm; the current location includes inside the smart refrigerator and/or outside the smart refrigerator;

其中，根据所述从属设备在所述终端设备中所处的当前位置，在所述从属设备中部署计算机视觉算法，包括：Wherein, deploying a computer vision algorithm in the slave device according to the current location of the slave device in the terminal device includes:

当所述智能摄像头所处的当前位置位于智能冰箱内部时，在该智能摄像头中部署物品识别算法；当所述智能摄像头所处的当前位置位于智能冰箱内部时，在该智能摄像头中部署人像识别算法。When the current location of the smart camera is inside the smart refrigerator, deploy an object recognition algorithm in the smart camera; when the current location of the smart camera is inside the smart refrigerator, deploy portrait recognition in the smart camera algorithm.

在本公开的一种示例性实施例中，控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，包括：In an exemplary embodiment of the present disclosure, controlling the slave device to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result includes:

主控设备在检测到与所述智能冰箱对应的第一预设位置存在目标对象时，根据主控设备与从属设备之间的当前设备映射关系表，获取处于该第一预设位置的从属设备与主控设备之间的通信地址；主控设备生成与处于第一预设位置的从属设备关联的人脸图像采集指令，并基于处于第一预设位置的从属设备与主控设备之间的通信地址，将所述人脸图像采集指令发送至处于第一预设位置的从属设备；处于第一预设位置的从属设备响应于所述人脸图像采集指令，采集所述目标对象的待识别人脸图像，并利用预设的人像识别算法对采集到的待识别人脸图像进行识别，得到人像识别结果。When the master control device detects that a target object exists at the first preset position corresponding to the smart refrigerator, it obtains the slave device at the first preset position according to the current device mapping relationship table between the master control device and the slave device. Communication address with the master control device; the master control device generates a face image collection instruction associated with the slave device at the first preset position, and based on the communication between the slave device and the master device at the first preset position Communication address, send the face image collection instruction to the slave device at the first preset position; the slave device at the first preset position responds to the face image collection instruction, collects the to-be-identified image of the target object Face images are collected, and the preset face recognition algorithm is used to recognize the collected face images to be recognized, and the face recognition results are obtained.

在本公开的一种示例性实施例中，利用预设的人像识别算法对采集到的待识别人脸图像进行识别，得到人像识别结果，包括：In an exemplary embodiment of the present disclosure, a preset portrait recognition algorithm is used to recognize the collected face image to be recognized, and a portrait recognition result is obtained, including:

利用预设的人脸检测与关键点定位工具，检测所述待识别人脸图像的待识别人脸区域；在所述待识别人脸区域中提取所述待识别人脸图像的待识别脸部关键点，并根据所述待识别脸部关键点，计算所述待识别人脸特征；计算所述待识别人脸特征以及处于第二预设位置处的从属设备中的特征值库中的原始人脸特征之间的欧式距离；根据所述欧式距离，得到所述待识别人脸特征与所述原始人脸特征之间的相似度比对结果，并根据所述相似度比对结果得到所述人像识别结果。Use preset face detection and key point positioning tools to detect the face area to be recognized in the face image to be recognized; extract the face to be recognized in the face image to be recognized in the face area to be recognized key points, and calculate the facial features to be recognized based on the key points of the face to be recognized; calculate the facial features to be recognized and the original values in the feature value library in the slave device at the second preset position Euclidean distance between facial features; according to the Euclidean distance, the similarity comparison result between the facial feature to be identified and the original facial feature is obtained, and the similarity comparison result is obtained based on the similarity comparison result Describe the portrait recognition results.

在本公开的一种示例性实施例中，控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，包括：In an exemplary embodiment of the present disclosure, controlling the slave device to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result includes:

主控设备在检测到与所述智能冰箱对应的第二预设位置存在新增物品时，根据主控设备与从属设备之间的当前设备映射关系表，获取处于该第二预设位置的从属设备与主控设备之间的通信地址；主控设备生成与处于第二预设位置的从属设备关联的物品图像采集指令，并基于处于第二预设位置的从属设备与主控设备之间的通信地址，将所述物品图像采集指令发送至处于第二预设位置的从属设备；处于第二预设位置的从属设备响应于所述物品图像采集指令，采集所述新增物品的待识别物品图像，并利用预设的物品识别算法对采集到的待识别物品图像进行识别，得到物品图像识别结果。When the master control device detects that there is a new item in the second preset position corresponding to the smart refrigerator, it obtains the slave at the second preset position according to the current device mapping relationship table between the master control device and the slave device. The communication address between the device and the master control device; the master control device generates an item image collection instruction associated with the slave device at the second preset position, and based on the communication between the slave device and the master device at the second preset position The communication address sends the item image collection instruction to the slave device at the second preset position; the slave device at the second preset position responds to the item image collection instruction and collects the items to be identified of the newly added items. image, and use the preset item recognition algorithm to identify the collected image of the item to be identified, and obtain the item image recognition result.

上述设备间的网络适配装置中各模块的具体细节已经在对应的设备间的网络适配方法中进行了详细的描述，因此此处不再赘述。The specific details of each module in the above-mentioned inter-device network adaptation device have been described in detail in the corresponding inter-device network adaptation method, so they will not be described again here.

应当注意，尽管在上文详细描述中提及了用于动作执行的设备的若干模块或者单元， 但是这种划分并非强制性的。实际上，根据本公开的实施方式，上文描述的两个或更多模块或者单元的特征和功能可以在一个模块或者单元中具体化。反之，上文描述的一个模块或者单元的特征和功能可以进一步划分为由多个模块或者单元来具体化。It should be noted that although several modules or units of the device for action execution are mentioned in the above detailed description, But this division is not mandatory. In fact, according to embodiments of the present disclosure, the features and functions of two or more modules or units described above may be embodied in one module or unit. Conversely, the features and functions of one module or unit described above may be further divided into being embodied by multiple modules or units.

此外，尽管在附图中以特定顺序描述了本公开中方法的各个步骤，但是，这并非要求或者暗示必须按照该特定顺序来执行这些步骤，或是必须执行全部所示的步骤才能实现期望的结果。附加的或备选的，可以省略某些步骤，将多个步骤合并为一个步骤执行，以及/或者将一个步骤分解为多个步骤执行等。Furthermore, although various steps of the methods of the present disclosure are depicted in the drawings in a specific order, this does not require or imply that the steps must be performed in that specific order, or that all of the illustrated steps must be performed to achieve the desired results. result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step for execution, and/or one step may be decomposed into multiple steps for execution, etc.

在本公开的示例性实施例中，还提供了一种能够实现上述方法的电子设备。In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

所属技术领域的技术人员能够理解，本公开的各个方面可以实现为***、方法或程序产品。因此，本公开的各个方面可以具体实现为以下形式，即：完全的硬件实施方式、完全的软件实施方式(包括固件、微代码等)，或硬件和软件方面结合的实施方式，这里可以统称为“电路”、“模块”或“***”。Those skilled in the art will understand that various aspects of the present disclosure may be implemented as systems, methods, or program products. Therefore, various aspects of the present disclosure may be embodied in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which may be collectively referred to herein as "Circuit", "Module" or "System".

下面参照图12来描述根据本公开的这种实施方式的电子设备1200。图12显示的电子设备1200仅仅是一个示例，不应对本公开实施例的功能和使用范围带来任何限制。An electronic device 1200 according to this embodiment of the present disclosure is described below with reference to FIG. 12 . The electronic device 1200 shown in FIG. 12 is only an example and should not bring any limitations to the functions and usage scope of the embodiments of the present disclosure.

如图12所示，电子设备1200以通用计算设备的形式表现。电子设备1200的组件可以包括但不限于：上述至少一个处理单元1210、上述至少一个存储单元1220、连接不同***组件(包括存储单元1220和处理单元1210)的总线1230以及显示单元1240。As shown in Figure 12, electronic device 1200 is embodied in the form of a general computing device. The components of the electronic device 1200 may include, but are not limited to: the above-mentioned at least one processing unit 1210, the above-mentioned at least one storage unit 1220, a bus 1230 connecting different system components (including the storage unit 1220 and the processing unit 1210), and the display unit 1240.

其中，所述存储单元存储有程序代码，所述程序代码可以被所述处理单元1210执行，使得所述处理单元1210执行本说明书上述“示例性方法”部分中描述的根据本公开各种示例性实施方式的步骤。例如，所述处理单元1210可以执行如图1中所示的步骤S110：从与所述主控设备关联的物理端口的内核节点中获取接入的从属设备的网络节点；步骤S120：从所述网络节点中读取所述从属设备在接入所述主控设备时，所经由的主控设备的通用串行总线接口的接口编号；步骤S130：基于所述接口编号以及所述从属设备所具有的当前通信网段，生成所述主控设备与所述从属设备之间的通信地址。Wherein, the storage unit stores program code, and the program code can be executed by the processing unit 1210, so that the processing unit 1210 performs various exemplary methods according to the present disclosure described in the "Example Method" section of this specification. Implementation steps. For example, the processing unit 1210 may perform step S110 as shown in Figure 1: obtain the network node of the accessed slave device from the kernel node of the physical port associated with the master device; step S120: obtain the network node of the accessed slave device from the kernel node of the physical port associated with the master device; Step S120: The network node reads the interface number of the universal serial bus interface of the master device through which the slave device accesses the master device; Step S130: Based on the interface number and the slave device's The current communication network segment is used to generate a communication address between the master device and the slave device.

存储单元1220可以包括易失性存储单元形式的可读介质，例如随机存取存储单元(RAM)12201和/或高速缓存存储单元12202，还可以进一步包括只读存储单元(ROM)12203。The storage unit 1220 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 12201 and/or a cache storage unit 12202, and may further include a read-only storage unit (ROM) 12203.

存储单元1220还可以包括具有一组(至少一个)程序模块12205的程序/实用工具12204，这样的程序模块12205包括但不限于：操作***、一个或者多个应用程序、其它程序模块以及程序数据，这些示例中的每一个或某种组合中可能包括网络环境的实现。Storage unit 1220 may also include a program/utility 12204 having a set of (at least one) program modules 12205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, Each of these examples, or some combination, may include the implementation of a network environment.

总线1230可以为表示几类总线结构中的一种或多种，包括存储单元总线或者存储单元控制器、***总线、图形加速端口、处理单元或者使用多种总线结构中的任意总线结构的局域总线。Bus 1230 may be a local area representing one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or using any of a variety of bus structures. bus.

电子设备1200也可以与一个或多个外部设备1300(例如键盘、指向设备、蓝牙设备等)通信，还可与一个或者多个使得用户能与该电子设备1200交互的设备通信，和/ 或与使得该电子设备1200能与一个或多个其它计算设备进行通信的任何设备(例如路由器、调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口1250进行。并且，电子设备1200还可以通过网络适配器1260与一个或者多个网络(例如局域网(LAN)，广域网(WAN)和/或公共网络，例如因特网)通信。如图所示，网络适配器1260通过总线1230与电子设备1200的其它模块通信。应当明白，尽管图中未示出，可以结合电子设备1200使用其它硬件和/或软件模块，包括但不限于：微代码、设备驱动器、冗余处理单元、外部磁盘驱动阵列、RAID***、磁带驱动器以及数据备份存储***等。Electronic device 1200 may also communicate with one or more external devices 1300 (e.g., keyboard, pointing device, Bluetooth device, etc.), and may also communicate with one or more devices that enable a user to interact with electronic device 1200, and/ Or with any device (eg, router, modem, etc.) that enables the electronic device 1200 to communicate with one or more other computing devices. This communication may occur through an input/output (I/O) interface 1250. Furthermore, the electronic device 1200 may also communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) through the network adapter 1260. As shown, network adapter 1260 communicates with other modules of electronic device 1200 via bus 1230. It should be understood that, although not shown in the figures, other hardware and/or software modules may be used in conjunction with electronic device 1200, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives And data backup storage system, etc.

通过以上的实施方式的描述，本领域的技术人员易于理解，这里描述的示例实施方式可以通过软件实现，也可以通过软件结合必要的硬件的方式来实现。因此，根据本公开实施方式的技术方案可以以软件产品的形式体现出来，该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM，U盘，移动硬盘等)中或网络上，包括若干指令以使得一台计算设备(可以是个人计算机、服务器、终端装置、或者网络设备等)执行根据本公开实施方式的方法。Through the above description of the embodiments, those skilled in the art can easily understand that the example embodiments described here can be implemented by software, or can be implemented by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, U disk, mobile hard disk, etc.) or on the network , including several instructions to cause a computing device (which may be a personal computer, a server, a terminal device, a network device, etc.) to execute a method according to an embodiment of the present disclosure.

在本公开的示例性实施例中，还提供了一种计算机可读存储介质，其上存储有能够实现本说明书上述方法的程序产品。在一些可能的实施方式中，本公开的各个方面还可以实现为一种程序产品的形式，其包括程序代码，当所述程序产品在终端设备上运行时，所述程序代码用于使所述终端设备执行本说明书上述“示例性方法”部分中描述的根据本公开各种示例性实施方式的步骤。In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the method described above in this specification is stored. In some possible implementations, various aspects of the present disclosure can also be implemented in the form of a program product, which includes program code. When the program product is run on a terminal device, the program code is used to cause the The terminal device performs the steps according to various exemplary embodiments of the present disclosure described in the above "Example Method" section of this specification.

根据本公开的实施方式的用于实现上述方法的程序产品，其可以采用便携式紧凑盘只读存储器(CD-ROM)并包括程序代码，并可以在终端设备，例如个人电脑上运行。然而，本公开的程序产品不限于此，在本文件中，可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行***、装置或者器件使用或者与其结合使用。The program product for implementing the above method according to an embodiment of the present disclosure may adopt a portable compact disk read-only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

所述程序产品可以采用一个或多个可读介质的任意组合。可读介质可以是可读信号介质或者可读存储介质。可读存储介质例如可以为但不限于电、磁、光、电磁、红外线、或半导体的***、装置或器件，或者任意以上的组合。可读存储介质的更具体的例子(非穷举的列表)包括：具有一个或多个导线的电连接、便携式盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。The program product may take the form of any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: electrical connection with one or more conductors, portable disk, hard disk, random access memory (RAM), read only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.

计算机可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了可读程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。可读信号介质还可以是可读存储介质以外的任何可读介质，该可读介质可以发送、传播或者传输用于由指令执行***、装置或者器件使用或者与其结合使用的程序。A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave carrying readable program code therein. Such propagated data signals may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A readable signal medium may also be any readable medium other than a readable storage medium that can send, propagate, or transport the program for use by or in connection with an instruction execution system, apparatus, or device.

可读介质上包含的程序代码可以用任何适当的介质传输，包括但不限于无线、有线、 光缆、RF等等，或者上述的任意合适的组合。Program code embodied on a readable medium may be transmitted over any suitable medium, including but not limited to wireless, wireline, Fiber optic cable, RF, etc., or any suitable combination of the above.

可以以一种或多种程序设计语言的任意组合来编写用于执行本公开操作的程序代码，所述程序设计语言包括面向对象的程序设计语言—诸如Java、C++等，还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算设备上执行、部分地在用户设备上执行、作为一个独立的软件包执行、部分在用户计算设备上部分在远程计算设备上执行、或者完全在远程计算设备或服务器上执行。在涉及远程计算设备的情形中，远程计算设备可以通过任意种类的网络，包括局域网(LAN)或广域网(WAN)，连接到用户计算设备，或者，可以连接到外部计算设备(例如利用因特网服务提供商来通过因特网连接)。Program code for performing operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., as well as conventional procedural Programming language—such as "C" or a similar programming language. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server execute on. In situations involving remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device, such as provided by an Internet service. (business comes via Internet connection).

此外，上述附图仅是根据本公开示例性实施例的方法所包括的处理的示意性说明，而不是限制目的。易于理解，上述附图所示的处理并不表明或限制这些处理的时间顺序。另外，也易于理解，这些处理可以是例如在多个模块中同步或异步执行的。In addition, the above-mentioned drawings are only schematic illustrations of processes included in the methods according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal sequence of these processes. In addition, it is also easy to understand that these processes may be executed synchronously or asynchronously in multiple modules, for example.

本领域技术人员在考虑说明书及实践这里发明的发明后，将容易想到本公开的其他实施例。本申请旨在涵盖本公开的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本公开的一般性原理并包括本公开未发明的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的，本公开的真正范围和精神由权利要求指出。 Other embodiments of the disclosure will be readily apparent to those skilled in the art from consideration of the specification and practice of the invention herein. This application is intended to cover any variations, uses, or adaptations of the disclosure that follow the general principles of the disclosure and include common knowledge or customary technical means in the technical field not invented by the disclosure . It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

Claims (14)

1. 一种设备间的网络适配方法，其特征在于，配置于终端设备，所述终端设备中包括主控设备以及从属设备；所述设备间的网络适配方法包括：A network adaptation method between devices, characterized in that it is configured in a terminal device, and the terminal device includes a master control device and a slave device; the network adaptation method between devices includes:

   从与所述主控设备关联的物理端口的内核节点中获取接入的从属设备的网络节点；Obtain the network node of the accessed slave device from the kernel node of the physical port associated with the master device;

   从所述网络节点中读取所述从属设备在接入所述主控设备时，所经由的主控设备的通用串行总线接口的接口编号；Read from the network node the interface number of the universal serial bus interface of the master device through which the slave device accesses the master device;

   基于所述接口编号以及所述从属设备所具有的当前通信网段，生成所述主控设备与所述从属设备之间的通信地址。Based on the interface number and the current communication network segment of the slave device, a communication address between the master device and the slave device is generated.
2. 根据权利要求1所述的设备间的网络适配方法，其特征在于，在从与所述主控设备关联的物理端口的目录下，获取接入的从属设备的网络节点之前，所述设备间的网络适配方法还包括：The network adaptation method between devices according to claim 1, characterized in that, before obtaining the network node of the accessed slave device from the directory of the physical port associated with the master control device, the device Network adaptation methods also include:

   启动预设程序进程，并基于所述预设程序进程对所述主控设备的物理端口的内核节点进行轮询监测；Start a preset program process, and perform polling monitoring of the kernel node of the physical port of the main control device based on the preset program process;

   在监测到所述物理端口中有从属设备接入时，在所述内核节点中创建与接入的从属设备关联的网络节点。When it is detected that a slave device is connected to the physical port, a network node associated with the connected slave device is created in the kernel node.
3. 根据权利要求2所述的设备间的网络适配方法，其特征在于，所述设备间的网络适配方法还包括：The network adaptation method between devices according to claim 2, characterized in that the network adaptation method between devices further includes:

   在所述主控设备中建立与接入的从属设备关联的虚拟节点，并计算已经建立的虚拟节点的节点数量；Establish a virtual node associated with the accessed slave device in the master control device, and calculate the number of established virtual nodes;

   判断所述节点数量是否大于所述主控设备所具有的通用串行总线接口的接口数量；Determine whether the number of nodes is greater than the number of universal serial bus interfaces of the master control device;

   在确定所述节点数量大于等于所述接口数量时，终止所述从属设备的接入。When it is determined that the number of nodes is greater than or equal to the number of interfaces, access of the slave device is terminated.
4. 根据权利要求1所述的设备间的网络适配方法，其特征在于，所述设备间的网络适配方法还包括：The network adaptation method between devices according to claim 1, characterized in that the network adaptation method between devices further includes:

   根据所述通信地址建立所述主控设备与从属设备之间的通信连接，并基于所述通信连接，在所述主控设备以及从属设备之间进行数据传输和/或指令传输。A communication connection is established between the master device and the slave device according to the communication address, and based on the communication connection, data transmission and/or instruction transmission is performed between the master device and the slave device.
5. 根据权利要求4所述的设备间的网络适配方法，其特征在于，所述设备间的网络适配方法还包括：The network adaptation method between devices according to claim 4, characterized in that the network adaptation method between devices further includes:

   根据所述通信地址，建立所述主控设备的通用串行总线接口与所述从属设备之间的当前设备映射关系表；According to the communication address, establish a current device mapping relationship table between the universal serial bus interface of the master device and the slave device;

   判断所述当前设备映射关系表与历史设备映射关系表之间是否存在新接入的从属设备；Determine whether there is a newly accessed slave device between the current device mapping relationship table and the historical device mapping relationship table;

   若存在新接入的从属设备，则为新接入的从属设备增加路由路径，若不存在新接入的从属设备，则对所述历史设备映射关系表中所包括的所有路由路径进行刷新。If there is a newly accessed slave device, a routing path is added for the newly accessed slave device. If there is no newly accessed slave device, all routing paths included in the historical device mapping relationship table are refreshed.
6. 根据权利要求5所述的设备间的网络适配方法，其特征在于，所述设备间的网络 适配方法还包括：The network adaptation method between devices according to claim 5, characterized in that the network between devices Adaptation methods also include:

   根据所述路由路径建立所述主控设备和/或从属设备与外部应用程序之间的通信连接。A communication connection between the master device and/or the slave device and an external application is established according to the routing path.
7. 根据权利要求4所述的设备间的网络适配方法，其特征在于，所述终端设备包括智能冰箱，所述主控设备包括主控SoC板卡，所述从属设备包括一个或者多个智能摄像头；The network adaptation method between devices according to claim 4, wherein the terminal device includes a smart refrigerator, the main control device includes a main control SoC board, and the slave device includes one or more smart cameras. ;

   其中，基于所述通信连接，在所述主控设备以及从属设备之间进行数据传输和/或指令传输，包括：Wherein, based on the communication connection, data transmission and/or instruction transmission is performed between the master control device and the slave device, including:

   根据所述从属设备在所述终端设备中所处的当前位置，在所述从属设备中部署计算机视觉算法；Deploying a computer vision algorithm in the slave device according to the current location of the slave device in the terminal device;

   控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，并基于所述通信连接图像识别结果从从属设备传输至主控设备。The slave device is controlled to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result, and the image recognition result is transmitted from the slave device to the master control device based on the communication connection.
8. 根据权利要求7所述的设备间的网络适配方法，其特征在于，所述计算机视觉算法包括人像识别算法和/或物品识别算法；所述当前位置包括智能冰箱内部和/或智能冰箱外部；The network adaptation method between devices according to claim 7, wherein the computer vision algorithm includes a portrait recognition algorithm and/or an object recognition algorithm; the current location includes the inside of the smart refrigerator and/or the outside of the smart refrigerator;

   其中，根据所述从属设备在所述终端设备中所处的当前位置，在所述从属设备中部署计算机视觉算法，包括：Wherein, deploying a computer vision algorithm in the slave device according to the current location of the slave device in the terminal device includes:

   当所述智能摄像头所处的当前位置位于智能冰箱内部时，在该智能摄像头中部署物品识别算法；When the current location of the smart camera is inside the smart refrigerator, deploy an object recognition algorithm in the smart camera;

   当所述智能摄像头所处的当前位置位于智能冰箱内部时，在该智能摄像头中部署人像识别算法。When the current location of the smart camera is inside the smart refrigerator, a portrait recognition algorithm is deployed in the smart camera.
9. 根据权利要求8所述的设备间的网络适配方法，其特征在于，控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，包括：The network adaptation method between devices according to claim 8, characterized in that controlling the slave device to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result includes:

   主控设备在检测到与所述智能冰箱对应的第一预设位置存在目标对象时，根据主控设备与从属设备之间的当前设备映射关系表，获取处于该第一预设位置的从属设备与主控设备之间的通信地址；When the master control device detects that a target object exists at the first preset position corresponding to the smart refrigerator, it obtains the slave device at the first preset position according to the current device mapping relationship table between the master control device and the slave device. Communication address with the main control device;

   主控设备生成与处于第一预设位置的从属设备关联的人脸图像采集指令，并基于处于第一预设位置的从属设备与主控设备之间的通信地址，将所述人脸图像采集指令发送至处于第一预设位置的从属设备；The master control device generates a face image collection instruction associated with the slave device at the first preset position, and collects the face image based on the communication address between the slave device at the first preset position and the master control device. sending the command to the slave device in the first preset position;

   处于第一预设位置的从属设备响应于所述人脸图像采集指令，采集所述目标对象的待识别人脸图像，并利用预设的人像识别算法对采集到的待识别人脸图像进行识别，得到人像识别结果。The slave device at the first preset position responds to the face image collection instruction, collects the face image to be recognized of the target object, and uses a preset portrait recognition algorithm to identify the collected face image to be recognized. , get the portrait recognition result.
10. 根据权利要求9所述的设备间的网络适配方法，其特征在于，利用预设的人像识别算法对采集到的待识别人脸图像进行识别，得到人像识别结果，包括：The network adaptation method between devices according to claim 9, characterized in that a preset portrait recognition algorithm is used to recognize the collected face images to be recognized, and the portrait recognition results are obtained, including:

    利用预设的人脸检测与关键点定位工具，检测所述待识别人脸图像的待识别人脸区域；Use preset face detection and key point positioning tools to detect the face area to be recognized in the face image to be recognized;

    在所述待识别人脸区域中提取所述待识别人脸图像的待识别脸部关键点，并根据所述待识别脸部关键点，计算所述待识别人脸特征； Extract the face key points to be recognized of the face image to be recognized in the face area to be recognized, and calculate the face features to be recognized based on the face key points to be recognized;

    计算所述待识别人脸特征以及处于第二预设位置处的从属设备中的特征值库中的原始人脸特征之间的欧式距离；Calculate the Euclidean distance between the facial feature to be recognized and the original facial feature in the feature value library in the slave device at the second preset position;

    根据所述欧式距离，得到所述待识别人脸特征与所述原始人脸特征之间的相似度比对结果，并根据所述相似度比对结果得到所述人像识别结果。According to the Euclidean distance, a similarity comparison result between the facial feature to be recognized and the original facial feature is obtained, and the portrait recognition result is obtained based on the similarity comparison result.
11. 根据权利要求8所述的设备间的网络适配方法，其特征在于，控制所述从属设备基于所述计算机视觉算法对采集到的待识别图像进行识别得到图像识别结果，包括：The network adaptation method between devices according to claim 8, characterized in that controlling the slave device to recognize the collected image to be recognized based on the computer vision algorithm to obtain an image recognition result includes:

    主控设备在检测到与所述智能冰箱对应的第二预设位置存在新增物品时，根据主控设备与从属设备之间的当前设备映射关系表，获取处于该第二预设位置的从属设备与主控设备之间的通信地址；When the master control device detects that there is a new item in the second preset position corresponding to the smart refrigerator, it obtains the slave at the second preset position according to the current device mapping relationship table between the master control device and the slave device. Communication address between the device and the main control device;

    主控设备生成与处于第二预设位置的从属设备关联的物品图像采集指令，并基于处于第二预设位置的从属设备与主控设备之间的通信地址，将所述物品图像采集指令发送至处于第二预设位置的从属设备；The master control device generates an item image collection instruction associated with the slave device at the second preset position, and sends the item image collection instruction based on the communication address between the slave device at the second preset position and the master control device. to the slave device in the second preset position;

    处于第二预设位置的从属设备响应于所述物品图像采集指令，采集所述新增物品的待识别物品图像，并利用预设的物品识别算法对采集到的待识别物品图像进行识别，得到物品图像识别结果。The slave device at the second preset position responds to the item image collection instruction, collects the item image to be identified of the newly added item, and uses the preset item recognition algorithm to identify the collected item image to be identified, and obtains Item image recognition results.
12. 一种设备间的网络适配装置，其特征在于，配置于终端设备，所述终端设备中包括主控设备以及从属设备；所述设备间的网络适配装置包括：A network adaptation device between devices, characterized in that it is configured in a terminal device, and the terminal device includes a master control device and a slave device; the network adaptation device between devices includes:

    网络节点获取模块，用于从与所述主控设备关联的物理端口的内核节点中获取接入的从属设备的网络节点；A network node acquisition module, configured to acquire the network node of the accessed slave device from the kernel node of the physical port associated with the master control device;

    接口编号读取模块，用于从所述网络节点中读取所述从属设备在接入所述主控设备时，所经由的主控设备的通用串行总线接口的接口编号；An interface number reading module, configured to read from the network node the interface number of the universal serial bus interface of the master device through which the slave device accesses the master device;

    通信地址生成模块，用于基于所述接口编号以及所述从属设备所具有的当前通信网段，生成所述主控设备与所述从属设备之间的通信地址。A communication address generation module, configured to generate a communication address between the master device and the slave device based on the interface number and the current communication network segment of the slave device.
13. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现权利要求1-11任一项所述的设备间的网络适配方法。A computer-readable storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, the network adaptation method between devices described in any one of claims 1-11 is implemented.
14. 一种电子设备，其特征在于，包括：An electronic device, characterized by including:

    处理器；以及processor; and

    存储器，用于存储所述处理器的可执行指令；memory for storing executable instructions for the processor;

    其中，所述处理器配置为经由执行所述可执行指令来执行权利要求1-11任一项所述的设备间的网络适配方法。 Wherein, the processor is configured to execute the network adaptation method between devices according to any one of claims 1-11 by executing the executable instructions.