WO2019179442A1

WO2019179442A1 - Interaction target determination method and apparatus for intelligent device

Info

Publication number: WO2019179442A1
Application number: PCT/CN2019/078748
Authority: WO
Inventors: 周子傲; 谢长武; 王雪松; 马健
Original assignee: 北京猎户星空科技有限公司
Priority date: 2018-03-21
Filing date: 2019-03-19
Publication date: 2019-09-26
Also published as: TW201941099A; CN108733208A

Abstract

The present invention relates to an interaction target determination method and apparatus for an intelligent device. The method comprises: acquiring an environment image in a monitoring range of an intelligent device, carrying out target identification on the environment image, using targets identified out from the environment image as candidate targets, acquiring state information of the candidate targets, for each candidate target, determining, according to the corresponding state information, whether the candidate target has an interaction intention of interacting with the intelligent device, and selecting an interaction target of the intelligent device from the candidate targets with the interaction intention. In this embodiment, by screening out the candidate targets with the interaction intention from all the candidate targets according to the state information of the candidate targets and further selecting the interaction target from the candidate targets with the interaction intention for the intelligent device, the purpose that the selected interaction target is the most possible one to be a target with the interaction intention with the intelligent device is achieved, the case of using a target without the interaction intention as the interaction target is avoided, and false startup of the intelligent device is reduced.

Description

智能设备的交互目标确定方法和装置Method and device for determining interactive target of intelligent device

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

本公开要求北京猎户星空科技有限公司于2018年03月21日提交的、发明名称为“智能设备的交互目标确定方法和装置”的、中国专利申请号“201810236768.7”的优先权。The present disclosure claims the priority of the Chinese Patent Application No. "201810236768.7" filed on March 21, 2018 by the Beijing Orion Star Technology Co., Ltd., which is entitled "Identification Target Method and Apparatus for Intelligent Devices".

技术领域Technical field

本公开涉及智能设备技术领域，尤其涉及一种智能设备的交互目标确定方法和装置。The present disclosure relates to the field of smart device technologies, and in particular, to a method and device for determining an interaction target of a smart device.

背景技术Background technique

随着智能设备技术的发展，已实现智能设备与人主动进行交互的过程。具体地，智能设备对一定范围的对象进行检测。当检测到人脸时，将人确定为交互目标，进行启动并主动与人进行交互。With the development of smart device technology, the process of intelligent device and human interaction has been realized. Specifically, the smart device detects a range of objects. When a human face is detected, the person is determined to be an interactive target, activated and actively interacts with the person.

发明内容Summary of the invention

本公开提出一种智能设备的交互目标确定方法，实现了从候选目标中筛除没有交互意图的目标，进而从存在交互意图的目标中筛选出交互目标，避免了将没有交互意图的目标选为交互目标，提高了交互目标的确定准确度，减少了智能设备的误启动。The present disclosure proposes a method for determining an interactive target of a smart device, which removes a target with no interaction intention from the candidate target, and then selects an interaction target from the target with the interaction intention, thereby avoiding selecting a target with no interaction intention as The interaction goal improves the accuracy of the determination of the interaction target and reduces the false start of the smart device.

本公开一方面实施例提出了一种智能设备的交互目标确定方法，包括：An embodiment of the present disclosure provides a method for determining an interaction target of a smart device, including:

获取在智能设备的监控范围内的环境图像，对所述环境图像进行目标识别；Obtaining an environment image within a monitoring range of the smart device, and performing target recognition on the environment image;

将从所述环境图像中识别出的目标作为候选目标，获取所述候选目标的状态信息；Obtaining the target information of the candidate target from the target identified in the environment image as a candidate target;

针对每个候选目标，根据对应的状态信息，判断是否存在与所述智能设备交互的交互意图；Determining, for each candidate target, whether there is an interaction intention of interacting with the smart device according to the corresponding state information;

从存在交互意图的候选目标中选取所述智能设备的交互目标。The interaction target of the smart device is selected from candidate objects with interaction intentions.

作为本公开一方面实施例一种可能的实现方式，所述获取所述候选目标的状态信息，包括：As a possible implementation manner of the embodiment of the present disclosure, the acquiring the status information of the candidate target includes:

获取所述候选目标与所述智能设备之间的距离；Obtaining a distance between the candidate target and the smart device;

针对每个候选目标，根据对应的状态信息，判断是否存在与所述智能设备交互的交互意图，包括：For each candidate target, determining, according to the corresponding state information, whether there is an interaction intention of interacting with the smart device, including:

针对每个候选目标，判断所述候选目标与所述智能设备之间的距离是否小于或者等于预设的所述距离阈值，且在所述距离阈值范围内的停留时长是否超出预设的时间阈值；Determining, for each candidate target, whether a distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether a stay duration within the distance threshold exceeds a preset time threshold ;

如果所述候选目标与所述智能设备之间的距离小于或者等于所述距离阈值且所述停留时长超出所述时间阈值，则确定所述该候选目标存在与所述智能设备交互的交互意图。If the distance between the candidate target and the smart device is less than or equal to the distance threshold and the stay duration exceeds the time threshold, determining that the candidate target has an interaction intention of interacting with the smart device.

获取所述候选目标与所述智能设备之间的距离，以及所述候选目标的人脸角度；Obtaining a distance between the candidate target and the smart device, and a face angle of the candidate target;

针对每个候选目标，根据对应的状态信息，判断是否存在与智能设备交互的交互意图，包括：For each candidate target, according to the corresponding status information, it is determined whether there is an interaction intention of interacting with the smart device, including:

针对每个候选目标，判断所述候选目标与所述智能设备之间的距离是否小于等于预设的距离阈值，且所述候选目标的人脸角度是否处于预设的角度范围内；Determining, for each candidate target, whether the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether the face angle of the candidate target is within a preset angle range;

如果所述候选目标与所述智能设备之间的距离小于或者等于预设的距离阈值，且所述候选目标的人脸角度处于预设的角度范围内，则确定所述候选目标存在与所述智能设备交互的交互意图。If the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and the face angle of the candidate target is within a preset angle range, determining that the candidate target exists and The interactive intent of the smart device interaction.

作为本公开一方面实施例一种可能的实现方式，从存在交互意图的所述候选目标中选取所述智能设备的交互目标，包括：As a possible implementation manner of an embodiment of the present disclosure, selecting an interaction target of the smart device from the candidate target having an interaction intention includes:

当检测到多个候选目标时，且存在交互意图的候选目标为多个时，从多个存在交互意图的所述候选目标中，确定出与所述智能设备距离最近的候选目标；When a plurality of candidate targets are detected, and there are a plurality of candidate targets having an interaction intention, determining, from the plurality of candidate targets having the interaction intention, a candidate target that is closest to the smart device;

从所述与所述智能设备距离最近的候选目标中，选取所述智能设备的交互目标。And selecting an interaction target of the smart device from the candidate target that is closest to the smart device.

作为本公开一方面实施例一种可能的实现方式，从所述与所述智能设备距离最近的候选目标中，选取所述智能设备的交互目标，包括：As a possible implementation manner of the embodiment of the present disclosure, selecting the interaction target of the smart device from the candidate target that is closest to the smart device includes:

当与所述智能设备距离最近的候选目标为多个时，查询所述智能设备的已注册用户人脸图像库中是否存在与所述智能设备距离最近的候选目标的人脸图像；When there are a plurality of candidate targets that are closest to the smart device, query whether there is a face image of the candidate target that is closest to the smart device in the registered user face image library of the smart device;

如果所述人脸图像库中存在一个与所述智能设备距离最近的候选目标的人脸图像，则将所述一个与所述智能设备距离最近的候选目标作为交互目标；If there is a face image of a candidate target that is closest to the smart device in the face image library, the candidate target that is closest to the smart device is used as an interaction target;

如果所述人脸图像库中不存在与所述智能设备距离最近的候选目标的人脸图像，则随机选取一个与所述智能设备距离最近的候选目标作为交互目标；If there is no face image of the candidate target closest to the smart device in the face image library, randomly selecting a candidate target that is closest to the smart device as an interaction target;

如果所述人脸图像库中存在多个与所述智能设备距离最近的候选目标的人脸图像，则将最先查询出的与所述智能设备距离最近的候选目标作为交互目标。If there are a plurality of face images of the candidate target that are closest to the smart device in the face image library, the candidate target that is the closest to the smart device that is firstly queried is used as the interaction target.

作为本公开一方面实施例一种可能的实现方式，所述获取所述候选目标与所述智能设备人之间的距离，包括：As a possible implementation manner of the embodiment of the present disclosure, the obtaining the distance between the candidate target and the smart device includes:

通过所述智能设备中的深度摄像头获取深度图，根据所述深度图，获取所述目标与所述智能设备之间的距离；或者，Obtaining a depth map by using a depth camera in the smart device, and acquiring a distance between the target and the smart device according to the depth map; or

通过所述智能设备中的双目视觉摄像头，对所述候选目标进行拍摄，计算所述双目视觉摄像头所拍摄图像的视差，根据所述视差计算所述候选目标与所述智能设备之间的距离；或者，Obtaining, by the binocular vision camera in the smart device, the candidate target, calculating a disparity of the image captured by the binocular vision camera, and calculating, between the candidate target and the smart device, according to the disparity Distance; or,

通过所述智能设备中的激光雷达，向所述监控范围内发射激光；Laser light is emitted into the monitoring range by a laser radar in the smart device;

根据处于所述监控范围内的每个障碍物返回的激光，生成每个障碍物的二值图；Generating a binary map of each obstacle based on the laser light returned by each obstacle within the monitored range;

将每个二值图与所述环境图像进行融合，从所有的二值图中识别出与所述候选目标对应的二值图；Combining each binary image with the environment image, and identifying a binary image corresponding to the candidate target from all the binary images;

根据所述候选目标对应的二值图的激光返回时间，确定出所述候选目标与所述智能设备之间的距离。Determining a distance between the candidate target and the smart device according to a laser return time of the binary image corresponding to the candidate target.

作为本公开一方面实施例一种可能的实现方式，所述获取所述候选目标的人脸角度，包括：As a possible implementation manner of the embodiment of the present disclosure, the acquiring a face angle of the candidate target includes:

从所述环境图像中截取所述候选目标的人脸图像；Extracting a face image of the candidate target from the environment image;

将所述人脸图像输入预先训练好的机器学习模型中，获取所述人脸图像中人脸的人脸角度；Inputting the face image into a pre-trained machine learning model, and acquiring a face angle of the face in the face image;

所述方法还包括：采用如下方式训练所述机器学习模型：The method also includes training the machine learning model in the following manner:

采集携带样本人脸图像，其中，所述样本人脸图像中携带标注数据，所述标注数据用于表示样本人脸的人脸角度；Collecting a sample face image, wherein the sample face image carries annotation data, and the annotation data is used to represent a face angle of the sample face;

将所述样本人脸图像输入到初始构建的机器学习模型中进行训练，当训练后的所述机器学习模型的误差在预设的误差范围内时，则得到训练好的所述机器学习模型。The sample face image is input into the initially constructed machine learning model for training, and when the error of the trained machine learning model is within a preset error range, the trained machine learning model is obtained.

作为本公开一方面实施例一种可能的实现方式，所述从存在交互意图的候选目标中选取所述智能设备的交互目标之后，还包括：As a possible implementation manner of the embodiment of the present disclosure, after selecting the interaction target of the smart device from the candidate target having the interaction intention, the method further includes:

控制所述智能设备与所述交互目标进行交互；Controlling the smart device to interact with the interaction target;

在交互过程中，识别所述交互目标的人脸图像的中心点；Identifying a center point of the face image of the interaction target during the interaction;

检测所述人脸图像的中心点是否在预设的图像区域内；Detecting whether a center point of the face image is within a preset image area;

如果未在所述图像区域内，获取所述人脸图像的中心点到达所述图像区域的中心点之间的路径；If not in the image area, acquiring a path between a center point of the face image and a center point of the image area;

根据所述路径，控制所述智能设备，使所述人脸图像的中心点在所述图像区域内。According to the path, the smart device is controlled such that a center point of the face image is within the image area.

本公开实施例的智能设备的交互目标确定方法，通过获取在智能设备的监控范围内的环境图像，对环境图像进行目标识别，将从环境图像中识别出的目标作为候选目标，获取候选目标的状态信息，针对每个候选目标，根据对应的状态信息，判断是否存在与智能设备交互的交互意图，从存在交互意图的候选目标中选取智能设备的交互目标。本实施例中，通过根据候选目标的状态信息，从所有候选目标中，筛选出存在交互意图的候选目标，进一步从存在交互意图的候选目标中，为智能设备选取出交互目标，使得选取的交互目标最可能是与智能设备有交互意图的目标，避免了将没有交互意图的目标作为交互目标，提高了交互目标的确定准确度，减少了智能设备的误启动。The method for determining an interaction target of the smart device according to the embodiment of the present disclosure, by acquiring an environment image within the monitoring range of the smart device, performing target recognition on the environment image, using the target identified from the environment image as a candidate target, and acquiring the candidate target The status information is used to determine, according to the corresponding status information, whether there is an interaction intention of interacting with the smart device, and select an interaction target of the smart device from the candidate targets having the interaction intention. In this embodiment, by selecting the candidate target with the interaction intention from all the candidate targets according to the state information of the candidate target, and further selecting the interaction target for the smart device from the candidate target having the interaction intention, so that the selected interaction is performed. The target is most likely to have a goal of interacting with the smart device, avoiding the goal of having no interaction intention as the interaction target, improving the accuracy of the determination of the interaction target, and reducing the false start of the smart device.

本公开另一方面实施例提出了一种智能设备的交互目标确定装置，包括：An embodiment of the present disclosure provides an interaction target determining apparatus for a smart device, including:

第一获取模块，用于在从存在交互意图的候选目标中选取智能设备的交互目标之后，获取在智能设备的监控范围内的环境图像，对所述环境图像进行目标识别；a first acquiring module, configured to acquire an environment image within a monitoring range of the smart device after the interaction target of the smart device is selected from the candidate target having the interaction intention, and perform target recognition on the environment image;

第二获取模块，用于将从所述环境图像中识别出的目标作为候选目标，获取所述候选目标的状态信息；a second acquiring module, configured to acquire, as a candidate target, a target object that is identified from the environment image, and acquire state information of the candidate target;

判断模块，用于针对每个候选目标，根据对应的状态信息，判断是否存在与所述智能设备交互的交互意图；a determining module, configured to determine, according to the corresponding state information, whether there is an interaction intention of interacting with the smart device according to each candidate target;

选取模块，用于从存在交互意图的候选目标中选取所述智能设备的交互目标。And a selection module, configured to select an interaction target of the smart device from a candidate target having an interaction intention.

作为本公开另一方面实施例一种可能的实现方式，所述第二获取模块具体用于：As a possible implementation manner of another embodiment of the present disclosure, the second acquiring module is specifically configured to:

所述判断模块具体用于：The determining module is specifically configured to:

针对每个候选目标，判断所述候选目标与所述智能设备之间的距离是否小于或者等于预设的距离阈值，且所述候选目标的人脸角度是否处于预设的角度范围内；Determining, for each candidate target, whether the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether the face angle of the candidate target is within a preset angle range;

作为本公开另一方面实施例一种可能的实现方式，所述选取模块包括：As a possible implementation manner of another embodiment of the present disclosure, the selecting module includes:

确定单元，用于当检测到多个候选目标时，且存在交互意图的候选目标为多个时，从多个存在交互意图的所述候选目标中，确定出与所述智能设备距离最近的候选目标；a determining unit, configured to: when a plurality of candidate targets are detected, and there are a plurality of candidate targets having an interaction intention, determine, from the plurality of candidate targets having the interaction intention, a candidate closest to the smart device aims;

选取单元，用于从所述与所述智能设备距离最近的候选目标中，选取所述智能设备的交互目标。And a selecting unit, configured to select an interaction target of the smart device from the candidate target that is closest to the smart device.

作为本公开另一方面实施例一种可能的实现方式，所述选取单元具体用于：As a possible implementation manner of another embodiment of the present disclosure, the selecting unit is specifically configured to:

将所述人脸图像输入预先训练好的机器学习模型中，获取所述人脸图像中人脸的人脸角度；Inputting the face image into a pre-trained machine learning model to obtain a face angle of a face in the face image;

所述装置还包括：The device also includes:

采集模块，用于采集携带样本人脸图像，其中，所述样本人脸图像中携带标注数据，所述标注数据用于表示样本人脸的人脸角度；An acquisition module, configured to collect a face image of the sample, wherein the sample face image carries the annotation data, and the annotation data is used to represent a face angle of the sample face;

训练模块，用于将所述样本人脸图像输入到初始构建的机器学习模型中进行训练，当训练后的所述机器学习模型的误差在预设的误差范围内时，则得到训练好的所述机器学习模型。a training module, configured to input the sample face image into an initially constructed machine learning model for training, and when the error of the machine learning model after training is within a preset error range, obtain a trained training A machine learning model.

作为本公开另一方面实施例一种可能的实现方式，所述装置还包括：As a possible implementation manner of another embodiment of the present disclosure, the device further includes:

第一控制模块，用于控制所述智能设备与所述交互目标进行交互；a first control module, configured to control the smart device to interact with the interaction target;

识别模块，用于在交互过程中，识别所述交互目标的人脸图像的中心点；a recognition module, configured to identify a center point of the face image of the interaction target during the interaction process;

检测模块，用于检测所述人脸图像的中心点是否在预设的图像区域内；a detecting module, configured to detect whether a center point of the face image is within a preset image area;

第三获取模块，用于在未在所述图像区域内时，获取所述人脸图像的中心点到达所述图像区域的中心点之间的路径；a third acquiring module, configured to acquire a path between a center point of the face image and a center point of the image area when not in the image area;

第二控制模块，用于根据所述路径，控制所述智能设备，使所述人脸图像的中心点在所述图像区域内。And a second control module, configured to control the smart device according to the path, so that a center point of the face image is within the image area.

本公开实施例的智能设备的交互目标确定装置，通过获取在智能设备的监控范围内的环境图像，对环境图像进行目标识别，将从环境图像中识别出的目标作为候选目标，获取候选目标的状态信息，针对每个候选目标，根据对应的状态信息，判断是否存在与智能设备交互的交互意图，从存在交互意图的候选目标中选取智能设备的交互目标。本实施例中，通过根据候选目标的状态信息，从所有候选目标中，筛选出存在交互意图的候选目标，进一步从存在交互意图的候选目标中，为智能设备选取出交互目标，使得选取的交互目标最可能是与智能设备有交互意图的目标，避免了将没有交互意图的目标作为交互目标，提高了交互目标的确定准确度，减少了智能设备的误启动。The interaction target determining apparatus of the smart device of the embodiment of the present invention performs target recognition on the environment image by acquiring an environment image within the monitoring range of the smart device, and uses the target identified from the environment image as a candidate target to acquire the candidate target. The status information is used to determine, according to the corresponding status information, whether there is an interaction intention of interacting with the smart device, and select an interaction target of the smart device from the candidate targets having the interaction intention. In this embodiment, by selecting the candidate target with the interaction intention from all the candidate targets according to the state information of the candidate target, and further selecting the interaction target for the smart device from the candidate target having the interaction intention, so that the selected interaction is performed. The target is most likely to have a goal of interacting with the smart device, avoiding the goal of having no interaction intention as the interaction target, improving the accuracy of the determination of the interaction target, and reducing the false start of the smart device.

本公开又一方面实施例提出了一种智能设备，包括：壳体、处理器、存储器、电路板和电源电路，其中，所述电路板安置在所述壳体围成的空间内部，所述处理器和所述存储器设置在所述电路板上；所述电源电路，用于为上述智能设备的各个电路或器件供电；所述存储器用于存储可执行程序代码；其中，所述处理器通过读取所述存储器中存储的可执行程序代码来运行与所述可执行程序代码对应的程序，以用于实现上述一方面实施例所述的智能设备的交互目标确定方法。A further aspect of the present disclosure provides a smart device, including: a housing, a processor, a memory, a circuit board, and a power supply circuit, wherein the circuit board is disposed inside a space enclosed by the housing, a processor and the memory are disposed on the circuit board; the power circuit is configured to supply power to each circuit or device of the smart device; the memory is configured to store executable program code; wherein the processor passes The executable program code stored in the memory is read to run a program corresponding to the executable program code for implementing the interactive target determining method of the smart device described in the above aspect.

本公开又一方面实施例提出了一种计算机程序产品，当所述计算机程序产品中的指令由处理器执行时实现如上述一方面实施例所述的智能设备的交互目标确定方法。A further aspect of the present disclosure provides a computer program product, which, when executed by a processor, implements an interactive object determination method of the smart device as described in the above-described embodiment.

本公开又一方面实施例提出了一种非临时性计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如上述一方面实施例所述的智能设备的交互目标确定方法。A further aspect of the present disclosure provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by the processor, implements an interactive target determination of the smart device as described in the above-described embodiment. method.

本公开附加的方面和优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本公开的实践了解到。The aspects and advantages of the present invention will be set forth in part in the description which follows.

附图说明DRAWINGS

本公开上述的和/或附加的方面和优点从下面结合附图对实施例的描述中将变得明显和容易理解，其中：The above and/or additional aspects and advantages of the present disclosure will become apparent and readily understood from

图1为本公开实施例提供的一种智能设备的交互目标确定方法的流程示意图；FIG. 1 is a schematic flowchart of a method for determining an interaction target of a smart device according to an embodiment of the present disclosure;

图2为本公开实施例提供的另一种智能设备的交互目标确定方法的流程示意图；FIG. 2 is a schematic flowchart diagram of another method for determining an interaction target of a smart device according to an embodiment of the present disclosure;

图3为本公开实施例提供的双目视觉计算距离的原理示意图；FIG. 3 is a schematic diagram of the principle of calculating binocular vision distance according to an embodiment of the present disclosure; FIG.

图4为本公开实施例提供的另一种智能设备的交互目标确定方法的流程示意图；FIG. 4 is a schematic flowchart diagram of another method for determining an interaction target of a smart device according to an embodiment of the present disclosure;

图5为本公开实施例提供的另一种智能设备的交互目标确定方法的流程示意图；FIG. 5 is a schematic flowchart of a method for determining an interaction target of another smart device according to an embodiment of the present disclosure;

图6为本公开实施例提出的另一种智能设备的交互目标确定方法的流程示意图；FIG. 6 is a schematic flowchart diagram of another method for determining an interaction target of a smart device according to an embodiment of the present disclosure;

图7为本公开实施例提供的一种智能设备的交互目标确定装置的结构示意图；FIG. 7 is a schematic structural diagram of an interaction target determining apparatus of a smart device according to an embodiment of the present disclosure;

图8为本公开智能设备一个实施例的结构示意图。FIG. 8 is a schematic structural diagram of an embodiment of a smart device according to the present disclosure.

具体实施方式detailed description

下面详细描述本公开的实施例，所述实施例的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例性的，旨在用于解释本公开，而不能理解为对本公开的限制。The embodiments of the present disclosure are described in detail below, and the examples of the embodiments are illustrated in the drawings, wherein the same or similar reference numerals are used to refer to the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the drawings are illustrative, and are not intended to be construed as limiting.

下面参考附图描述本公开实施例的智能设备的交互目标确定方法和装置。The method and apparatus for determining an interaction target of a smart device according to an embodiment of the present disclosure will be described below with reference to the accompanying drawings.

本公开各实施例，针对智能设备在检测到人脸时，将人作为交互目标的确定交互目标的方法，可能会将与智能设备没有交互意愿的目标作为交互目标，从而造成智能设备误启动的问题，提出一种智能设备的交互目标确定方法。In various embodiments of the present disclosure, a method for determining an interaction target by using a person as an interaction target when a smart device detects a human face may use a target that does not have a willingness to interact with the smart device as an interaction target, thereby causing the smart device to be manually activated. The problem is to propose a method for determining the interaction target of a smart device.

本公开实施例的智能设备的交互目标确定方法，通过根据候选目标的状态信息，从所有候选目标中，筛选出存在交互意图的候选目标，进一步从存在交互意图的候选目标中，为智能设备选取出交互目标，使得选取的交互目标最可能是与智能设备有交互意图的目标，避免了将没有交互意图的目标作为交互目标，提高了交互目标的确定准确度，减少了智能设备的误启动。The interaction target determining method of the smart device of the embodiment of the present disclosure selects candidate objects having interaction intentions from all the candidate targets according to the state information of the candidate targets, and further selects the smart objects from the candidate targets with the interaction intentions. The interaction target is selected so that the selected interaction target is most likely to have a goal of interacting with the smart device, avoiding the goal of having no interaction intention as the interaction target, improving the determination accuracy of the interaction target, and reducing the false start of the smart device.

图1为本公开实施例提供的一种智能设备的交互目标确定方法的流程示意图。FIG. 1 is a schematic flowchart diagram of a method for determining an interaction target of a smart device according to an embodiment of the present disclosure.

如图1所示，该智能设备的交互目标确定方法包括：As shown in FIG. 1 , the method for determining an interaction target of the smart device includes:

步骤101，获取在智能设备的监控范围内的环境图像，对环境图像进行目标识别。Step 101: Acquire an environment image within a monitoring range of the smart device, and perform target recognition on the environment image.

本实施例中，智能设备可以是机器人、智能家电等。In this embodiment, the smart device may be a robot, a smart home appliance, or the like.

智能设备上配置有摄像装置，如摄像头，智能设备通过摄像装置可实时采集监控范围内的环境图像。在获取环境图像后，可对环境图像进行检测，以识别进入监控范围的目标。其中，这里的目标可以理解为人。The smart device is equipped with a camera device, such as a camera, and the smart device can collect the environment image in the monitoring range in real time through the camera device. After the environment image is acquired, the environment image can be detected to identify the target that enters the monitoring range. Among them, the goal here can be understood as human.

以识别环境图像中的人为例，智能设备可通过人脸检测或者人体检测，识别环境图像中的人。具体而言，从环境图像中提取物体的轮廓，将提取的物体轮廓与预存的人脸轮廓或人体轮廓，进行比对。当提取的轮廓与预设的轮廓之间的相似度超过预设的阈值，可以认为从环境图像中识别到了人。从而，通过该方法可以识别出环境图像中所有的人。Taking the person in the environment image as an example, the smart device can recognize the person in the environment image through face detection or human body detection. Specifically, the outline of the object is extracted from the environment image, and the extracted object outline is compared with the pre-existing face contour or the human body contour. When the similarity between the extracted contour and the preset contour exceeds a preset threshold, it can be considered that a person is recognized from the environmental image. Thus, all people in the environmental image can be identified by this method.

步骤102，将从环境图像中识别出的目标作为候选目标，获取候选目标的状态信息。Step 102: The target information identified by the environment image is used as a candidate target, and the state information of the candidate target is acquired.

本实施例中，当从环境图像中识别出目标时，将识别出的目标作为候选目标。例如，当有人进入机器人的监控范围内时，机器人可从采集的环境图像中识别出进入监控范围内的人，这些人均作为候选目标。In the present embodiment, when the target is recognized from the environment image, the recognized target is taken as the candidate target. For example, when someone enters the monitoring range of the robot, the robot can identify the people entering the monitoring range from the collected environmental images, and these people are candidates.

在识别出环境图像中的目标后，获取候选目标的状态信息，如目标的位置、目标在距离阈值范围内的停留时间、在预设时长内目标被识别到的次数等，以根据候选目标的状态信息，确定候选目标是否存在与智能设备交互的交互意图。After identifying the target in the environment image, acquiring state information of the candidate target, such as the location of the target, the dwell time of the target within the distance threshold, the number of times the target is recognized within the preset duration, etc., according to the candidate target The status information determines whether the candidate target has an interaction intent to interact with the smart device.

步骤103，针对每个候选目标，根据对应的状态信息，判断是否存在与智能设备交互的交互意图。Step 103: Determine, for each candidate target, whether there is an interaction intention of interacting with the smart device according to the corresponding state information.

相关技术中，智能设备在识别出人脸之后，直接将人作为交互目标，与人交互。但是，智能设备识别出的人可能与智能设备没有交互的意愿，由此可能会造成误启动。In the related art, after the smart device recognizes the face, the smart person directly interacts with the person as an interaction target. However, the person identified by the smart device may not have the willingness to interact with the smart device, which may cause a false start.

本实施例中，针对每个候选目标，根据候选目标的状态信息，判断候选目标是否存在交互意图。In this embodiment, for each candidate target, whether the candidate target has an interaction intention is determined according to the state information of the candidate target.

作为一种可能的实现方式，获取预设时长内候选目标被识别到的次数，并将该次数与预设的次数进行比较。如果预设时长内目标被识别到的次数，大于预设的次数，可以认为目标经常出现，与智能设备之间存在交互意图。As a possible implementation manner, the number of times the candidate target is recognized within the preset duration is obtained, and the number of times is compared with the preset number of times. If the number of times the target is recognized within the preset duration is greater than the preset number of times, the target may be considered to appear frequently, and there is an interaction intention with the smart device.

例如，在过去一个月内，公司前台的机器人识别到某人的次数为4次，大于预设的次数2次，说明这个人是公司的常客，可以确定该人与机器人之间存在交互意图。For example, in the past month, the robot at the front desk of the company recognized the number of times of a person to 4 times, which is greater than the preset number of times, indicating that the person is a frequent visitor of the company and can determine that there is an interaction intention between the person and the robot.

本实施例中，根据候选目标的状态信息，从候选目标中筛除没有交互意图的候选目标，从而可以避免将没有交互意图的目标，确定为交互目标。In this embodiment, according to the state information of the candidate target, the candidate target without the interaction intention is screened out from the candidate target, so that the target without the interaction intention can be avoided from being determined as the interaction target.

步骤104，从存在交互意图的候选目标中选取智能设备的交互目标。Step 104: Select an interaction target of the smart device from the candidate targets having the interaction intention.

本实施例中，为了进一步提高确定交互目标的准确性，降低智能设备误启动的概率，可继续从存在交互意图的候选目标中选取智能设备的交互目标，从而使交互目标为最可能存在交互意图的候选目标。In this embodiment, in order to further improve the accuracy of determining the interaction target and reduce the probability of the smart device being falsely activated, the interaction target of the smart device may be selected from the candidate targets with the interaction intention, so that the interaction target is the most likely to have the interaction intention. Candidate target.

如果存在交互意图的候选目标仅有一个，则将该候选目标作为交互目标。如果存在交互意图的目标有多个时，可以根据候选目标与智能设备之间的距离，确定交互目标。具体过程后续实施例，将进行详细说明，在此不再赘述。If there is only one candidate target with the interaction intention, the candidate target is used as the interaction target. If there are multiple targets with interaction intentions, the interaction target can be determined according to the distance between the candidate target and the smart device. Specific embodiments of the specific processes will be described in detail, and are not described herein again.

在上述实施例的基础上，对于步骤103，根据对应的状态信息，判断是否存在与所述智能设备交互的交互意图，作为一种可能的实现方式，可根据候选目标与智能设备之间的距离，和候选目标在预设的距离阈值范围内的停留时间，判断候选目标是否存在交互意图。图2为本公开实施例提供的另一种智能设备的交互目标确定方法的流程示意图。On the basis of the foregoing embodiment, for step 103, it is determined whether there is an interaction intention of interacting with the smart device according to the corresponding state information. As a possible implementation manner, the distance between the candidate target and the smart device may be determined. And the dwell time of the candidate target within a preset distance threshold range to determine whether the candidate target has an interaction intention. FIG. 2 is a schematic flowchart diagram of another method for determining an interaction target of a smart device according to an embodiment of the present disclosure.

如图2所示，该智能设备的交互目标确定方法包括：As shown in FIG. 2, the method for determining an interaction target of the smart device includes:

步骤201，获取在智能设备的监控范围内的环境图像，对环境图像进行目标识别。Step 201: Acquire an environment image within a monitoring range of the smart device, and perform target recognition on the environment image.

本实施例中，智能设备获取监控范围内的环境图像，以及环境图像进行目标识别的方法，可参见上述实施例中记载的相关内容，在此不再赘述。In this embodiment, the method for obtaining the environment image in the monitoring range and the method for the object image to be recognized by the smart device may be referred to the related content described in the foregoing embodiment, and details are not described herein again.

步骤202，将从环境图像中识别出的目标作为候选目标，获取候选目标与智能设备之间的距离。Step 202: The target identified from the environment image is used as a candidate target, and the distance between the candidate target and the smart device is obtained.

可以理解的是，候选目标与智能设备之间的距离越近，说明候选目标与智能设备之间存在交互意图的可能性越大，因此本实施例中，将候选目标与智能设备之间的距离，作为判断候选目标是否存在，与智能设备交互的交互意图的依据之一。It can be understood that the closer the distance between the candidate target and the smart device is, the more likely the interaction intention exists between the candidate target and the smart device. Therefore, in this embodiment, the distance between the candidate target and the smart device is As one of the basis for judging whether the candidate target exists or not, the interaction intention of interacting with the smart device.

本实施例中，可通过深度摄像头或者双目视觉摄像头或者激光雷达，获取候选目标与智能设备之间的距离。In this embodiment, the distance between the candidate target and the smart device can be obtained by a depth camera or a binocular vision camera or a laser radar.

作为一种可能的实现方式，智能设备中配置有深度摄像头，通过深度摄像头，获取候选目标的深度图。在具体实现时，可通过结构光投射器向候选目标表面投射可控制的光点、光条或光面结构，并由深度摄像头中的图像传感器获得图像，通过几何关系，利用三角原理计算得到候选目标的三维坐标，从而可以得到候选目标与智能设备之间的距离。As a possible implementation manner, the smart device is configured with a depth camera, and the depth map of the candidate target is obtained through the depth camera. In a specific implementation, a controllable light spot, a light strip or a smooth surface structure can be projected to the candidate target surface by the structured light projector, and an image is obtained by the image sensor in the depth camera, and the candidate is calculated by using the triangular principle through the geometric relationship. The three-dimensional coordinates of the target, so that the distance between the candidate target and the smart device can be obtained.

作为另一种可能的实现方式，在智能设备中配置双目视觉摄像头，通过双目视觉摄像头，对候选目标进行拍摄。然后，计算双目视觉摄像头所拍摄图像的视差，根据视差计算候选目标与智能设备之间的距离。As another possible implementation, a binocular vision camera is configured in the smart device, and the candidate target is captured by the binocular vision camera. Then, the parallax of the image captured by the binocular vision camera is calculated, and the distance between the candidate target and the smart device is calculated based on the parallax.

图3为本公开实施例提供的双目视觉计算距离的原理示意图。图3中，在实际空间中，画出了两个摄像头所在位置O _l和O _r，以及左右摄像头的光轴线，两个摄像头的焦平面，焦平面距离两个摄像头所在平面的距离为f。 FIG. 3 is a schematic diagram of the principle of calculating binocular vision distance according to an embodiment of the present disclosure. In Fig. 3, in the actual space, the positions O _l and O _{r of} the two cameras are drawn, and the optical axes of the left and right cameras, the focal planes of the two cameras, and the focal plane are at a distance f from the plane of the two cameras.

如图3所示，p和p′分别是同一候选目标P在不同拍摄图像中的位置。其中，p点距离所在拍摄图像的左侧边界的距离为x _l，p′点距离所在拍摄图像的左侧边界的距离为x _r。O _l和O _r分别为两个摄像头，这两个摄像头在同一平面，两个摄像头之间的距离为Z。 As shown in FIG. 3, p and p' are the positions of the same candidate target P in different captured images, respectively. Wherein, the distance from the p-point to the left boundary of the captured image is x _l , and the distance from the p-point to the left boundary of the captured image is x _r . O _l and _Or are respectively two cameras, the two cameras are in the same plane, and the distance between the two cameras is Z.

基于三角测距原理，图3中的P与两个摄像头所在平面之间的距离b，具有如下关系：

Based on the principle of triangulation, the distance b between P in Figure 3 and the plane of the two cameras has the following relationship:

基于此，可以推得

其中，d为同一候选目标双目摄像头所拍摄图像的视觉差。由于Z、f为定值，因此，根据视觉差d可以确定出候选目标与摄像头所在平面之间的距离b，即候选目标与智能设备之间的距离。 Based on this, you can push

Where d is the visual difference of the image captured by the same candidate target binocular camera. Since Z and f are constant values, the distance b between the candidate target and the plane of the camera, that is, the distance between the candidate target and the smart device, can be determined according to the visual difference d.

作为再一种可能的实现方式，在智能设备中配置激光雷达，通过激光雷达向监控范围内发射激光，发射的激光遇到监控范围内的障碍物将被反射。智能设备接收监控范围内的每个障碍物返回的激光，根据返回的激光生成每个障碍物的二值图。然后，将每个二值图与环境图像进行融合，从所有二值图中识别出与候选目标对应的二值图。具体地，可以根据每个障碍物的二值图可以识别出每个障碍物的轮廓或者大小，然后将环境图像中每个目标的轮廓或者大小进行匹配，从而可以得到候选目标对应的二值图。之后，将候选目标对应的二值图的激光返回时间乘以光速，并除以2，得到候选目标与智能设备之间的距离。As a further possible implementation, the laser radar is arranged in the smart device, and the laser is emitted into the monitoring range by the laser radar, and the emitted laser encounters obstacles within the monitoring range to be reflected. The smart device receives the laser returned by each obstacle within the monitored range and generates a binary map of each obstacle based on the returned laser. Then, each binary image is fused with the environment image, and the binary image corresponding to the candidate target is identified from all the binary images. Specifically, the contour or size of each obstacle can be identified according to the binary map of each obstacle, and then the contour or size of each target in the environment image is matched, so that the binary map corresponding to the candidate target can be obtained. . Then, the laser return time of the binary image corresponding to the candidate target is multiplied by the speed of light, and divided by 2 to obtain the distance between the candidate target and the smart device.

需要说明的是，其他用于计算候选目标与智能设备之间的距离的方法，也包含在本公开实施例的保护范围内。It should be noted that other methods for calculating the distance between the candidate target and the smart device are also included in the protection scope of the embodiments of the present disclosure.

步骤203，针对每个候选目标，判断候选目标与智能设备之间的距离是否小于或者等于预设的距离阈值，且在距离阈值范围内的停留时长是否超出预设的时间阈值。Step 203: Determine, for each candidate target, whether the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether the stay duration within the distance threshold exceeds a preset time threshold.

由于当候选目标与智能设备之间的距离较远时，候选目标可能不存在与智能设备交互的交互意图，或者距离较近，但候选目标的停留时间较短也可能不存在与智能设备交互的交互意图。Because when the distance between the candidate target and the smart device is far, the candidate target may not have the interaction intention of interacting with the smart device, or the distance is relatively close, but the short stay time of the candidate target may not exist to interact with the smart device. Interactive intent.

由此，可针对每个候选目标，将候选目标与智能设备之间的距离，与预设的距离阈值进行比较，以判断选目标与智能设备之间的距离是否小于或者等于预设的距离阈值。如果距离在距离阈值范围内，判断候选目标在距离阈值范围内停留的时间是否超过预设的时间阈值。Therefore, for each candidate target, the distance between the candidate target and the smart device can be compared with a preset distance threshold to determine whether the distance between the selected target and the smart device is less than or equal to a preset distance threshold. . If the distance is within the distance threshold, it is determined whether the candidate target stays within the distance threshold exceeds a preset time threshold.

步骤204，如果距离小于或者等于距离阈值且停留时长超出时间阈值，则确定该候选目标存在与智能设备交互的交互意图。Step 204: If the distance is less than or equal to the distance threshold and the stay duration exceeds the time threshold, determine that the candidate target has an interaction intention of interacting with the smart device.

当候选目标与智能设备之间的距离小于预设的距离阈值，且候选目标在距离阈值范围内的停留时长超过预设的时间阈值，可以认为候选目标存在与智能设备交互的交互意图。When the distance between the candidate target and the smart device is less than the preset distance threshold, and the staying duration of the candidate target within the distance threshold exceeds the preset time threshold, the candidate target may be considered to have an interaction intention of interacting with the smart device.

以机器人为例，若人与机器人之间的距离小于3米，且人在3米内停留的时间超过2秒，可以认为人存在与机器人交互的交互意图。Taking the robot as an example, if the distance between the person and the robot is less than 3 meters, and the person stays within 3 meters for more than 2 seconds, it can be considered that the person has an interactive intention of interacting with the robot.

步骤205，从存在交互意图的候选目标中选取智能设备的交互目标。Step 205: Select an interaction target of the smart device from the candidate targets having the interaction intention.

本实施例中，步骤205与上述实施例中的步骤104类似，故在此不再赘述。In this embodiment, the step 205 is similar to the step 104 in the foregoing embodiment, and therefore is not described herein again.

本公开实施例的智能设备的交互目标确定方法，通过候选目标与智能设备之间的距离，以及候选目标在预设的距离阈值范围内的停留时间，从所有候选目标中，筛选出存在与智能设备交互的交互意图的候选目标，相比在检测到人脸时，直接将人作为交互目标，可以降低智能设备的误启动。The interaction target determining method of the smart device of the embodiment of the present disclosure selects presence and intelligence from all candidate targets by the distance between the candidate target and the smart device and the dwell time of the candidate target within a preset distance threshold range. The candidate target of the interaction intention of the device interaction can reduce the false start of the smart device by directly using the person as the interaction target when detecting the face.

对于步骤103，作为另一种可能的实现方式，也可根据候选目标与智能设备之间的距离，以及候选目标的人脸角度，判断候选目标是否存在与智能设备交互的交互意图。图4为本公开实施例提供的另一种智能设备的交互目标确定方法的流程示意图。For step 103, as another possible implementation manner, whether the candidate target has an interaction intention of interacting with the smart device may be determined according to the distance between the candidate target and the smart device and the face angle of the candidate target. FIG. 4 is a schematic flowchart diagram of another method for determining an interaction target of a smart device according to an embodiment of the present disclosure.

如图4所示，该智能设备的交互目标确定方法包括：As shown in FIG. 4, the method for determining an interaction target of the smart device includes:

步骤301，获取在智能设备的监控范围内的环境图像，对环境图像进行目标识别。Step 301: Acquire an environment image within a monitoring range of the smart device, and perform target recognition on the environment image.

本实施例中，步骤301与上述实施例中的步骤101类似，故在此不再赘述。In this embodiment, step 301 is similar to step 101 in the foregoing embodiment, and details are not described herein again.

步骤302，将从环境图像中识别出的目标作为候选目标，获取候选目标与智能设备之间的距离，以及候选目标的人脸角度。Step 302: The target identified from the environment image is used as a candidate target, and the distance between the candidate target and the smart device and the face angle of the candidate target are acquired.

在实际中，当人路过机器人时，如果人转头看向机器人，或者当人脸正对机器人时，说明人对机器人的关注度较高，人存在与机器人交互的交互意图。由此，可将人脸图像中人脸的人脸角度，作为判断候选目标是否存与智能设备交互的交互意图的依据之一。In practice, when a person passes by a robot, if a person turns his head to look at the robot, or when the face is facing the robot, it means that the person pays more attention to the robot, and the person has an interactive intention of interacting with the robot. Thereby, the face angle of the face in the face image can be used as one of the basis for determining whether the candidate target has an interaction intention of interacting with the smart device.

本实施例中，可通过候选目标与智能设备之间的距离，以及候选目标的人脸角度，来判断候选目标是否存在与智能设备交互的交互意愿。其中，在获取候选目标与智能设备之间的距离时，可通过上述实施例中的记载的方法获取。In this embodiment, whether the candidate target has an interactive willingness to interact with the smart device may be determined by the distance between the candidate target and the smart device and the face angle of the candidate target. Wherein, when the distance between the candidate target and the smart device is acquired, the method can be obtained by the method described in the above embodiment.

在获取人脸角度时，可通过预先训练好的机器学习模型，获取人脸角度。具体地，可按照人脸轮廓从环境图像中截取候选目标的人脸图像，之后将人脸图像输入到机器学习模型中。机器学习模型根据人脸图像，输出人脸图像中人脸角度。When acquiring the face angle, the face angle can be obtained through a pre-trained machine learning model. Specifically, the face image of the candidate target may be extracted from the environment image according to the face contour, and then the face image is input into the machine learning model. The machine learning model outputs the face angle in the face image according to the face image.

其中，人脸角度可以是人脸中轴线偏离人脸图像中轴线的角度，人脸中轴线包括水平方向的中轴线和垂直方向的中轴线，相应的人脸图像中轴线也包括水平方向的中轴线和垂直方向的中轴线。从人脸图像中可以识别出人脸水平方向中轴线和垂直方向的中轴线，分别偏离与人脸图像的水平方向的中轴线和人脸图像的垂直方向的中轴线的角度，获取到的角度就是人脸角度。The face angle may be an angle from which the central axis of the face deviates from the central axis of the face image, and the central axis of the face includes a central axis in the horizontal direction and a central axis in the vertical direction, and the central axis of the corresponding face image also includes the horizontal direction. The central axis of the axis and the vertical direction. The central axis of the horizontal direction of the face and the central axis of the vertical direction can be recognized from the face image, respectively, and are offset from the central axis of the horizontal direction of the face image and the central axis of the vertical direction of the face image, and the obtained angle is obtained. It is the angle of the face.

本实施例中，可采用如下方式训练机器学习模型。首先，采集人脸图像，并对人脸图像进行人脸角度标注，从而使样本人脸图像，携带表示样本人脸图像的人脸角度的标注数据。之后，将样本人脸图像输入到初始构建的机器学习模型中进行训练。当机器学习模型输出的人脸角度，与标注的人脸角度之间的差值，在预设的误差范围内时，可以认为机器学习模型已经训练完毕。In this embodiment, the machine learning model can be trained in the following manner. First, the face image is collected, and the face image is labeled with the face angle, so that the sample face image carries the annotation data of the face angle representing the sample face image. After that, the sample face image is input into the initially constructed machine learning model for training. When the difference between the face angle output by the machine learning model and the face angle of the annotation is within the preset error range, the machine learning model can be considered to have been trained.

本实施例中，通过训练好的机器学习模型获取人脸角度，可以提高获取的人脸角度的精度，从而能够提高后续判断的准确性。In this embodiment, by acquiring the face angle by the trained machine learning model, the accuracy of the acquired face angle can be improved, thereby improving the accuracy of the subsequent judgment.

步骤303，针对每个候选目标，判断候选目标与智能设备之间的距离是否小于或者等于预设的距离阈值，且候选目标的人脸角度是否处于预设的角度范围内。Step 303: Determine, for each candidate target, whether the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether the face angle of the candidate target is within a preset angle range.

本实施例中，针对每个候选目标，将候选目标与智能设备之间的距离，与预设的距离阈值进行比较，将候选目标的人脸角度与预设的角度范围的上限值进行比较。In this embodiment, for each candidate target, the distance between the candidate target and the smart device is compared with a preset distance threshold, and the face angle of the candidate target is compared with the upper limit of the preset angle range. .

假设距离阈值为3米，角度范围为[0°，45°]，判断候选目标与智能设备之间的距离是否小于3米，将人脸角度与45°进行比较，以判断人脸角度是否处于预设的角度范围内。Assume that the distance threshold is 3 meters and the angle range is [0°, 45°]. Determine whether the distance between the candidate target and the smart device is less than 3 meters, and compare the face angle with 45° to determine whether the face angle is at Within the preset angle range.

步骤304，如果候选目标与智能设备之间的距离小于或者等于预设的距离阈值，且候选目标的人脸角度处于预设的角度范围内，则确定候选目标存在与智能设备交互的交互意图。Step 304: If the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and the face angle of the candidate target is within a preset angle range, determining that the candidate target has an interaction intention of interacting with the smart device.

本实施例中，当候选目标与智能设备之间的距离小于或等于预设的距离阈值，并且候选目标的人脸角度处于预设的角度范围内，说明候选目标在距离阈值范围内，对智能设备进行关注，可以确定候选目标存在与智能设备交互的交互意图。相比直接将检测到人作为交互目标而言，提高了交互目标确认的准确度。In this embodiment, when the distance between the candidate target and the smart device is less than or equal to the preset distance threshold, and the face angle of the candidate target is within a preset angle range, indicating that the candidate target is within the distance threshold, the smart The device pays attention to determine that the candidate target has an interaction intention of interacting with the smart device. The accuracy of the cross-target confirmation is improved compared to directly detecting the person as an interaction target.

步骤305，从存在交互意图的候选目标中选取智能设备的交互目标。Step 305: Select an interaction target of the smart device from the candidate targets having the interaction intention.

本实施例中，步骤305与上述实施例中步骤104类似，在此不再赘述。In this embodiment, step 305 is similar to step 104 in the foregoing embodiment, and details are not described herein again.

本公开实施例的智能设备的交互目标确定方法，通过候选目标与智能设备之间的距离，以及候选目标人脸角度，从所有候选目标中，筛选出存在与智能设备交互的交互意图的候选目标，相比在检测到人脸时，直接将人作为交互目标，可以降低智能设备的误启动。In the interaction target determining method of the smart device of the embodiment of the present disclosure, the candidate target having the interaction intention of interacting with the smart device is selected from all the candidate targets by the distance between the candidate target and the smart device and the candidate target face angle Compared with detecting a human face, directly using a person as an interactive target can reduce the false start of the smart device.

需要说明的是，在判断候选目标是否存在交互意图时，也可在候选目标与智能设备之间的距离，在距离阈值范围内，且候选目标在距离阈值范围内停留的时间超过时间阈值，以及候选目标的人脸角度在预设的角度范围内时，确定候选目标存在与智能设备交互的交互意图。否则，可以认为候选目标不存在与智能设备交互的交互意图。It should be noted that, when determining whether the candidate target has an interaction intention, the distance between the candidate target and the smart device may be within the distance threshold, and the candidate target stays within the distance threshold exceeds the time threshold, and When the face angle of the candidate target is within a preset angle range, it is determined that the candidate target has an interaction intention of interacting with the smart device. Otherwise, the candidate target may be considered to have no interactive intent to interact with the smart device.

上述实施例中，对于从存在交互意图的候选目标中选取智能设备的交互目标，当存在交互意图的候选目标只有一个时，可将存在交互意图的候选目标作为交互目标。当存在交互意图的候选目标有多个时，可根据候选目标与智能设备之间的距离，确定从候选目标中选取交互目标。图5为本公开实施例提供的另一种智能设备的交互目标确定方法的流程示意图。In the above embodiment, for selecting an interaction target of the smart device from the candidate targets having the interaction intention, when there is only one candidate target having the interaction intention, the candidate target having the interaction intention may be the interaction target. When there are multiple candidate targets with interaction intentions, the interaction target may be selected from the candidate targets according to the distance between the candidate target and the smart device. FIG. 5 is a schematic flowchart diagram of another method for determining an interaction target of a smart device according to an embodiment of the present disclosure.

如图5所示，对于步骤104，该智能设备的交互目标确定方法可包括：As shown in FIG. 5, for step 104, the interaction target determining method of the smart device may include:

步骤401，当检测到多个候选目标时，且存在交互意图的候选目标为多个时，从多个存在交互意图的候选目标中，确定出与智能设备距离最近的候选目标。Step 401: When multiple candidate targets are detected, and there are multiple candidate targets with interaction intentions, the candidate targets closest to the smart device are determined from the plurality of candidate targets with the interaction intention.

由于候选目标与智能设备之间的距离越近，说明候选目标与智能设备之间的交互意图越强。The closer the distance between the candidate target and the smart device, the stronger the interaction intention between the candidate target and the smart device.

本实施例中，当智能设备从环境图像中检测到多个候选目标，且判断出存在交互意图的候选目标也为多个时，可将多个存在交互意图的候选目标与智能设备之间的距离进行比较，以从多个存在交互意图的候选目标中，查找出与智能设备距离最近的候选目标，从而筛选出交互意图较强的候选目标。In this embodiment, when the smart device detects a plurality of candidate targets from the environment image, and determines that there are multiple candidate targets having the interaction intention, the plurality of candidate targets having the interaction intention may be between the smart target device and the smart device. The distance is compared to find a candidate target that is closest to the smart device from the plurality of candidate targets with the interaction intention, thereby filtering out the candidate target with strong interaction intention.

步骤402，从与智能设备距离最近的候选目标中，选取智能设备的交互目标。Step 402: Select an interaction target of the smart device from a candidate target that is closest to the smart device.

本实施例中，为了进一步确定智能设备的交互目标，需要从与智能设备距离最近的目标中，选取智能设备的交互目标。In this embodiment, in order to further determine the interaction target of the smart device, the interaction target of the smart device needs to be selected from the target that is closest to the smart device.

可以理解的是，当与智能设备距离最近的候选目标仅有一个时，可将该候选目标作为智能设备的交互目标。It can be understood that when there is only one candidate target closest to the smart device, the candidate target can be used as the interaction target of the smart device.

当与智能设备距离最近的候选目标有多个时，需要从多个与智能设备距离最近的候选目标中，选取智能设备的交互目标。When there are multiple candidate targets that are closest to the smart device, the interaction target of the smart device needs to be selected from a plurality of candidate targets that are closest to the smart device.

以机器人为例，某公司前台放置一个机器人，当用户需要进入公司时，可以在机器人中进行信息登录，即在机器人中进行注册。或者可以从公司网站中下载注册用户的人脸图像，存储到机器人中，在公司网站中注册过的用户，同步地在机器人中进行了注册。一般在该机器人中注册过的用户，比未注册过的用户与机器人交互的交互意图更强。由此，可根据候选目标是否已注册，确定智能设备的交互目标。Take the robot as an example. A robot is placed in the foreground of a company. When the user needs to enter the company, the information can be registered in the robot, that is, registered in the robot. Alternatively, the face image of the registered user can be downloaded from the company website, stored in the robot, and the user registered in the company website is registered in the robot synchronously. Users who are registered in the robot generally have a stronger interaction intention than the unregistered user and the robot. Thus, the interaction target of the smart device can be determined according to whether the candidate target has been registered.

机器人在日常接待工作时，可以采集访客或者公司员工的人脸图像，利用采集的访客或者公司员工的人脸图像，构建一个已注册用户人脸图像库，也可以利用网站注册用户的人脸图像，构建该人脸图像库。The robot can collect the face image of the visitor or company employee during the daily reception work, construct a registered user face image library by using the captured face image of the visitor or company employee, or use the website to register the user's face image. , construct the face image library.

作为一种可能的实现方式，智能设备可在本地查询与智能设备距离最近的候选目标，是否已经注册智能设备。具体地，智能设备可预先存储已注册用户人脸图像库，人脸图像库中存储有已注册智能设备的用户人脸图像。当与智能设备距离最近的候选目标为多个时，可将与智能设备距离最近的候选目标的人脸图像，与人脸图像库中的人脸图像进行比较。As a possible implementation manner, the smart device can locally query the candidate target that is closest to the smart device, and whether the smart device has been registered. Specifically, the smart device may pre-store the registered user face image library, and the face image library stores the user face image of the registered smart device. When there are a plurality of candidate targets closest to the smart device, the face image of the candidate target closest to the smart device may be compared with the face image in the face image library.

如果人脸图像库中存一个与智能设备距离最近的候选目标的人脸图像，说明该候选目标已注册，那么将该候选目标作为智能设备的交互目标。If the face image of the candidate target that is closest to the smart device is stored in the face image library, indicating that the candidate target is registered, the candidate target is used as the interaction target of the smart device.

如果人脸图像库中不存在与智能设备距离最近的候选目标的人脸图像，说明与智能设备距离最近的候选目标均未注册，可从与智能设备距离最近的候选目标中，随机选取一个候选目标作为交互目标。If there is no face image of the candidate target closest to the smart device in the face image database, the candidate target closest to the smart device is not registered, and a candidate may be randomly selected from the candidate target closest to the smart device. The goal is the interaction goal.

如果人脸图像库中存在多个与智能设备距离最近的候选目标的人脸图像，说明有多个与智能设备距离最近的候选目标已注册，那么可将最先查询出的与智能设备距离最近的候选目标作为交互目标，也可从已注册且与智能设备距离最近的候选目标中，随机选取一个候选目标作为交互目标。If there are multiple face images of the candidate target closest to the smart device in the face image library, indicating that a plurality of candidate targets closest to the smart device are registered, the first query may be the closest to the smart device. The candidate target is used as an interaction target, and a candidate target can be randomly selected from the candidate targets that are registered and closest to the smart device as the interaction target.

作为另一种可能的实现方式，当与智能设备距离最近的候选目标为多个时，可将所有与智能设备距离最近的候选目标的人脸图像，发送给服务器，由服务器返回查询结果至智能设备，智能设备根据比较结果确定交互目标。As another possible implementation manner, when there are multiple candidate targets closest to the smart device, all face images of the candidate target closest to the smart device may be sent to the server, and the server returns the query result to the smart The device, the smart device determines the interaction target based on the comparison result.

具体地，服务器存储有已注册用户的人脸图像库，当与智能设备距离最近的候选目标为多个时，智能设备将多个与智能设备距离最近的候选目标的人脸图像，发送至服务器。服务器接收到人脸图像，并在已注册用户人脸图像库中，查询是否存在与智能设备距离最近的候选目标的人脸图像。然后，服务器将查询结果发送给智能设备。智能设备根据查询结果，确定智能设备的交互目标，具体的确定方法可参见上述方法，在此不再赘述。Specifically, the server stores a face image library of the registered user. When the number of candidate targets closest to the smart device is multiple, the smart device sends a plurality of face images of the candidate target closest to the smart device to the server. . The server receives the face image, and in the registered user face image library, queries whether there is a face image of the candidate target closest to the smart device. The server then sends the query results to the smart device. The smart device determines the interaction target of the smart device according to the query result. For the specific determination method, refer to the foregoing method, and details are not described herein again.

举例来说，A从机器人面前路过，并没有交互意图，而B是公司的常客，之前已经完成了注册。当A和B与机器人的距离小于距离阈值3米，且与机器人的距离相同时，机器人可选取已经完成的注册B作为交互目标，向B打招呼。For example, A passes by in front of the robot and has no intent to interact, while B is a frequent visitor to the company and has previously completed registration. When the distance between A and B and the robot is less than the distance threshold of 3 meters and the distance from the robot is the same, the robot can select the registered B that has been completed as the interactive target and greet B.

本公开实施例的智能设备的交互目标确定方法，在存在交互意图的候选目标有多个时，筛选出与智能设备距离最近的候选目标，在与智能设备距离最近的目标有多个时，通过查询已注册用户人脸图像库，根据查询结果，选取智能设备的交互目标，而相关技术中智能设备从同时出现的多人中，选取的交互目标可能并不是最可能与智能设备有交互意图的人，从而提高了交互目标的确定准确度，避免智能设备的误启动。The interaction target determining method of the smart device of the embodiment of the present disclosure selects a candidate target that is closest to the smart device when there are multiple candidate targets with the interaction intention, and when there are multiple targets closest to the smart device, Querying the registered user face image database, and selecting the interaction target of the smart device according to the query result, and the smart device selected from the multi-person at the same time in the related art may not be the most likely to interact with the smart device. People, thereby improving the accuracy of the determination of the interaction target, and avoiding the false start of the smart device.

在实际中，智能设备确定交互目标后，在智能设备与交互目标交互的过程中，交互目标可能处于移动状态，例如，机器人向交互目标打招呼的过程中，可能人处于移动状态。为了使智能设备达到保持正面跟随人交互特点，本公开实施例还提出，在交互过程中，使人脸图像的中心点处于图像区域内。图6为本公开实施例提出的另一种智能设备的交互目标确定方法的流程示意图。In practice, after the smart device determines the interaction target, the interaction target may be in a moving state during the interaction between the smart device and the interaction target. For example, in the process of the robot greeting the interactive target, the person may be in a moving state. In order to enable the smart device to maintain the positive follower interaction feature, the embodiment of the present disclosure further provides that the center point of the face image is within the image region during the interaction. FIG. 6 is a schematic flowchart diagram of another method for determining an interaction target of a smart device according to an embodiment of the present disclosure.

在从存在交互意图的候选目标中选取智能设备的交互目标之后，如图6所示，该智能设备的交互目标确定方法还包括：After the interaction target of the smart device is selected from the candidate target having the interaction intention, as shown in FIG. 6, the interaction target determination method of the smart device further includes:

步骤105，控制智能设备与交互目标进行交互。Step 105: Control the smart device to interact with the interaction target.

本实施例中，在确定交互目标后，智能设备启动，并与交互目标进行交互。以机器人为例，机器人在确定打招呼的对象后，启动并与交互目标打招呼，如“欢迎光临”。In this embodiment, after determining the interaction target, the smart device starts and interacts with the interaction target. Taking the robot as an example, after determining the object to greet, the robot starts and greets the interactive target, such as "Welcome."

步骤106，在交互过程中，识别交互目标的人脸图像的中心点。Step 106: In the interaction process, identify a center point of the face image of the interaction target.

其中，目标的人脸图像可以是环境图像中包含目标的人脸的最小区域的图像。The face image of the target may be an image of the smallest area of the face containing the target in the environment image.

本实施例中，在交互的过程，智能设备实时识别交互目标的人脸图像的中心点。其中，人脸图像的中心点是人脸图像的竖直中心线和水平中心线的交点。In this embodiment, in the process of interaction, the smart device identifies the center point of the face image of the interaction target in real time. Wherein, the center point of the face image is the intersection of the vertical center line of the face image and the horizontal center line.

步骤107，检测人脸图像的中心点是否在预设的图像区域内。Step 107: Detect whether a center point of the face image is within a preset image area.

本实施例中，预设的图像区域可以是以环境图像的中心点为圆心，以预设尺寸画圆，得到的圆形区域。其中，预设尺寸可以是人在距离阈值处时，人脸图像的水平尺寸的一半。当然，也可以根据需要进行设置。In this embodiment, the preset image area may be a circular area obtained by drawing a circle at a preset size with the center point of the environment image as a center. The preset size may be half of the horizontal size of the face image when the person is at the distance threshold. Of course, you can also set it as needed.

智能设备可每隔预设时间，如每隔0.5秒，检测人脸图像的中心点，是否在预设的图像区域内，以判断人脸图像是否在预设的图像区域内。The smart device may detect the center point of the face image every preset time, such as every 0.5 seconds, whether it is within the preset image area to determine whether the face image is within the preset image area.

步骤108，如果未在图像区域内，获取人脸图像的中心点到达图像区域的中心点之间的路径。Step 108: If not in the image area, obtain a path between the center point of the face image and the center point of the image area.

本实施例中，如果人脸图像的中心点未在图像区域内，说明智能设备能够捕捉到的人脸画面不够完整，则获取人脸图像的中心点到图像区域的中心点之间的路径。In this embodiment, if the center point of the face image is not in the image area, indicating that the face image that the smart device can capture is not complete, the path between the center point of the face image and the center point of the image area is acquired.

步骤109，根据路径控制智能设备，使人脸图像的中心点在图像区域内。Step 109: Control the smart device according to the path, so that the center point of the face image is within the image area.

在智能设备获取到人脸图像的中心点到图像区域的中心点之间的路径后，根据路径控制智能设备，使人脸图像的中心点在图像区域内。After the smart device acquires the path between the center point of the face image and the center point of the image area, the smart device is controlled according to the path so that the center point of the face image is within the image area.

作为一种可能的实现方式，可以图像区域的中心点为圆心，建立直角坐标系，获取人脸图像的中心点的坐标，并计算出人脸图像的中心点与图像区域的中心点之间的距离，以及人脸图像的中心点相对水平方向的夹角。之后，控制智能设备转动相应的角度和距离。As a possible implementation manner, the center point of the image area may be a center, a Cartesian coordinate system is established, coordinates of a center point of the face image are acquired, and a center point of the face image and a center point of the image area are calculated. The distance, and the angle between the center point of the face image and the horizontal direction. After that, the smart device is controlled to rotate the corresponding angle and distance.

以机器人为例，若机器人检测到人脸图像的中心点，在图像区域中心点的右侧，说明人逐渐向右移动，则控制机器人的云台、底盘向右转动，以对人进行跟随，实现注视的目的。Taking the robot as an example, if the robot detects the center point of the face image and the right side of the center point of the image area indicates that the person gradually moves to the right, the pan/tilt and the chassis of the control robot are rotated to the right to follow the person. To achieve the purpose of gaze.

本公开实施例的智能设备的交互目标确定方法，通过检测人脸图像的中心点是否在预设的图像区域内，实现智能设备对交互目标的跟随，使智能设备与人交互时更加生动、灵活。The method for determining the interaction target of the smart device in the embodiment of the present disclosure is to detect whether the center point of the face image is in the preset image area, so that the smart device can follow the interaction target, so that the smart device interacts with the person more vividly and flexibly. .

为了实现上述实施例，本公开实施例还提出一种智能设备的交互目标确定装置。图7为本公开实施例提供的一种智能设备的交互目标确定装置的结构示意图。In order to implement the foregoing embodiments, an embodiment of the present disclosure further provides an interaction target determining apparatus of a smart device. FIG. 7 is a schematic structural diagram of an interaction target determining apparatus of a smart device according to an embodiment of the present disclosure.

如图7所示，该智能设备的交互目标确定装置包括：第一获取模块510、第二获取模块520、判断模块530、选取模块540。As shown in FIG. 7, the interaction target determining apparatus of the smart device includes: a first acquiring module 510, a second obtaining module 520, a determining module 530, and a selecting module 540.

第一获取模块510用于获取在智能设备的监控范围内的环境图像，对环境图像进行目标识别。The first obtaining module 510 is configured to acquire an environment image within a monitoring range of the smart device, and perform target recognition on the environment image.

第二获取模块520用于将从环境图像中识别出的目标作为候选目标，获取候选目标的状态信息。The second obtaining module 520 is configured to acquire the state information of the candidate target by using the target identified from the environment image as a candidate target.

判断模块530用于针对每个候选目标，根据对应的状态信息，判断是否存在与智能设备交互的交互意图。The determining module 530 is configured to determine, according to the corresponding state information, whether there is an interaction intention of interacting with the smart device for each candidate target.

选取模块540用于从存在交互意图的候选目标中选取智能设备的交互目标。The selection module 540 is configured to select an interaction target of the smart device from the candidate targets having the interaction intention.

在本实施例一种可能的实现方式中，上述第二获取模块520具体用于：In a possible implementation manner of this embodiment, the foregoing second obtaining module 520 is specifically configured to:

获取候选目标与所述智能设备之间的距离；Obtaining a distance between the candidate target and the smart device;

上述判断模块530具体用于：The determining module 530 is specifically configured to:

针对每个候选目标，判断候选目标与智能设备之间的距离是否小于或者等于预设的距离阈值，且在距离阈值范围内的停留时长是否超出预设的时间阈值；Determining, for each candidate target, whether the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether the stay duration within the distance threshold exceeds a preset time threshold;

如果候选目标与智能设备之间的距离小于或者等于距离阈值且所述停留时长超出所述时间阈值，则确定该候选目标存在与智能设备交互的交互意图。If the distance between the candidate target and the smart device is less than or equal to the distance threshold and the stay duration exceeds the time threshold, it is determined that the candidate target has an interaction intent to interact with the smart device.

获取候选目标与智能设备之间的距离，以及候选目标的人脸角度；Obtaining the distance between the candidate target and the smart device, and the face angle of the candidate target;

针对每个候选目标，判断候选目标与智能设备之间的距离是否小于等于预设的距离阈值，且候选目标的人脸角度是否处于预设的角度范围内；Determining, for each candidate target, whether the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether the face angle of the candidate target is within a preset angle range;

如果候选目标与智能设备之间的距离小于或等于预设的距离阈值，且候选目标的人脸角度处于预设的角度范围内，则确定候选目标存在与智能设备交互的交互意图。If the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and the face angle of the candidate target is within a preset angle range, determining that the candidate target has an interaction intention of interacting with the smart device.

在本实施例一种可能的实现方式中，上述选取模块540包括：In a possible implementation manner of the embodiment, the selecting module 540 includes:

确定单元，用于当检测到多个候选目标时，且存在交互意图的候选目标为多个时，从多个存在交互意图的候选目标中，确定出与智能设备距离最近的候选目标；a determining unit, configured to: when a plurality of candidate targets are detected, and there are multiple candidate targets with interaction intentions, determine a candidate target that is closest to the smart device from the plurality of candidate targets having the interaction intention;

选取单元，用于从与智能设备距离最近的候选目标中，选取智能设备的交互目标。The selecting unit is configured to select an interaction target of the smart device from a candidate target that is closest to the smart device.

在本实施例一种可能的实现方式中，上述选取单元具体用于：In a possible implementation manner of this embodiment, the foregoing selecting unit is specifically configured to:

当与智能设备距离最近的候选目标为多个时，查询智能设备的已注册用户人脸图像库中是否存在与智能设备距离最近的候选目标的人脸图像；When there are a plurality of candidate targets that are closest to the smart device, whether there is a face image of the candidate target that is closest to the smart device in the registered user face image library of the smart device;

如果人脸图像库中存在一个与智能设备距离最近的候选目标的人脸图像，则将一个与智能设备距离最近的候选目标作为交互目标；If there is a face image of the candidate target closest to the smart device in the face image library, a candidate target closest to the smart device is used as the interaction target;

如果人脸图像库中不存在与智能设备距离最近的候选目标的人脸图像，则随机选取一个与智能设备距离最近的候选目标作为交互目标；If there is no face image of the candidate target closest to the smart device in the face image database, a candidate target closest to the smart device is randomly selected as the interaction target;

如果人脸图像库中存在多个与智能设备距离最近的候选目标的人脸图像，则将最先查询出的与智能设备距离最近的候选目标作为交互目标。If there are a plurality of face images of the candidate target that are closest to the smart device in the face image library, the candidate target that is the closest to the smart device is searched for as the interaction target.

通过智能设备中的深度摄像头获取深度图，根据深度图，获取目标与智能设备之间的距离；或者，Obtain a depth map through a depth camera in the smart device, and obtain a distance between the target and the smart device according to the depth map; or

通过智能设备中的双目视觉摄像头，对候选目标进行拍摄，计算双目视觉摄像头所拍摄图像的视差，根据视差计算候选目标与智能设备之间的距离；或者，The candidate target is captured by the binocular vision camera in the smart device, the parallax of the image captured by the binocular vision camera is calculated, and the distance between the candidate target and the smart device is calculated according to the parallax; or

通过智能设备中的激光雷达，向监控范围内发射激光；Laser light is emitted into the monitoring range by laser radar in the smart device;

根据处于监控范围内的每个障碍物返回的激光，生成每个障碍物的二值图；Generate a binary map of each obstacle based on the laser returned by each obstacle within the monitored range;

将每个二值图与环境图像进行融合，从所有的二值图中识别出与候选目标对应的二值图；Combining each binary image with the environment image, and identifying a binary image corresponding to the candidate target from all the binary images;

根据候选目标对应的二值图的激光返回时间，确定出候选目标与智能设备之间的距离。The distance between the candidate target and the smart device is determined according to the laser return time of the binary map corresponding to the candidate target.

从环境图像中截取候选目标的人脸图像；Obtaining a face image of the candidate target from the environment image;

将人脸图像输入预先训练好的机器学习模型中，获取人脸图像中人脸的人脸角度；Inputting a face image into a pre-trained machine learning model to obtain a face angle of a face in the face image;

在本实施例一种可能的实现方式中，所述装置还包括：In a possible implementation manner of this embodiment, the apparatus further includes:

采集模块，用于采集携带样本人脸图像，其中，样本人脸图像中携带标注数据，标注数据用于表示样本人脸的人脸角度；An acquisition module is configured to collect a face image of the sample, wherein the sample face image carries the annotation data, and the annotation data is used to represent the face angle of the sample face;

训练模块，用于将样本人脸图像输入到初始构建的机器学习模型中进行训练，当训练后的机器学习模型的误差在预设的误差范围内时，则得到训练好的机器学习模型。The training module is configured to input the sample face image into the initially constructed machine learning model for training, and when the error of the trained machine learning model is within a preset error range, the trained machine learning model is obtained.

在本实施例一种可能的实现方式中，该装置还包括：In a possible implementation manner of this embodiment, the apparatus further includes:

第一控制模块，用于在从存在交互意图的候选目标中选取智能设备的交互目标之后，控制智能设备与交互目标进行交互；a first control module, configured to control the smart device to interact with the interaction target after selecting the interaction target of the smart device from the candidate target having the interaction intention;

识别模块，用于在交互过程中，识别交互目标的人脸图像的中心点；a recognition module, configured to identify a center point of the face image of the interaction target during the interaction process;

检测模块，用于检测人脸图像的中心点是否在预设的图像区域内；a detecting module, configured to detect whether a center point of the face image is within a preset image area;

第三获取模块，用于在未在图像区域内时，获取人脸图像的中心点到达图像区域的中心点之间的路径；a third acquiring module, configured to acquire a path between a center point of the face image and a center point of the image area when not in the image area;

第二控制模块，用于根据路径，控制智能设备，使人脸图像的中心点在图像区域内。The second control module is configured to control the smart device according to the path, so that the center point of the face image is within the image area.

需要说明的是，前述对智能设备的交互目标确定方法实施例的解释说明，也适用于该实施例的智能设备的交互目标确定装置，故在此不再赘述。It should be noted that the foregoing description of the embodiment of the method for determining the interaction target of the smart device is also applicable to the interaction target determining device of the smart device of the embodiment, and therefore is not described herein again.

本公开实施例的智能设备的交互目标确定装置，通过获取在智能设备的监控范围内的环境图像，对环境图像进行目标识别，将从环境图像中识别出的目标作为候选目标，获取候选目标的状态信息，针对每个候选目标，根据对应的状态信息，判断是否存在与智能设备交互的交互意图，从存在交互意图的候选目标中选取智能设备的交互目标。本实施例中，通过根据候选目标的状态信息，从所有候选目标中，筛选出存在交互意图的候选目标，进一步从存在交互意图的候选目标中，为智能设备选取交互目标，使得选取的交互目标最可能是与智能设备有交互意图的目标，避免了将没有交互意图的目标作为交互目标，提高了交互目标的确定准确度，减少了智能设备的误启动。The interaction target determining apparatus of the smart device of the embodiment of the present invention performs target recognition on the environment image by acquiring an environment image within the monitoring range of the smart device, and uses the target identified from the environment image as a candidate target to acquire the candidate target. The status information is used to determine, according to the corresponding status information, whether there is an interaction intention of interacting with the smart device, and select an interaction target of the smart device from the candidate targets having the interaction intention. In this embodiment, the candidate target having the interaction intention is selected from all the candidate targets according to the state information of the candidate target, and the interaction target is selected for the smart device from the candidate target having the interaction intention, so that the selected interaction target is selected. It is most likely to have a goal of interacting with the smart device, avoiding the goal of having no interaction intention as the interaction target, improving the determination accuracy of the interaction target, and reducing the false start of the smart device.

为了实现上述实施例，本公开实施例还提出一种智能设备。In order to implement the above embodiments, an embodiment of the present disclosure further provides a smart device.

图8为本公开智能设备一个实施例的结构示意图，如图8所示，该智能设备可包括：壳体610、处理器620、存储器630、电路板640和电源电路650，其中，电路板640安置在壳体610围成的空间内部，处理器620和存储器630设置在电路板640上；电源电路650，用于为上述智能设备的各个电路或器件供电；存储器630用于存储可执行程序代码；处理器620通过读取存储器630中存储的可执行程序代码来运行与可执行程序代码对应的程序，用于执行上述实施例所述的智能设备的交互目标确定方法。FIG. 8 is a schematic structural diagram of an embodiment of a smart device according to the present disclosure. As shown in FIG. 8 , the smart device may include a housing 610 , a processor 620 , a memory 630 , a circuit board 640 , and a power circuit 650 . The circuit board 640 . Placed inside the space enclosed by the housing 610, the processor 620 and the memory 630 are disposed on the circuit board 640; the power supply circuit 650 is configured to supply power to the respective circuits or devices of the smart device; and the memory 630 is configured to store executable program code. The processor 620 runs a program corresponding to the executable program code by reading the executable program code stored in the memory 630 for executing the interactive target determining method of the smart device described in the above embodiments.

为了实现上述实施例，本公开实施例还提出一种计算机程序产品，当计算机程序产品中的指令由处理器执行时实现如上述实施例所述的智能设备的交互目标确定方法。In order to implement the above embodiments, the embodiment of the present disclosure further provides a computer program product, which implements an interactive target determining method of the smart device as described in the foregoing embodiments when the instructions in the computer program product are executed by the processor.

为了实现上述实施例，本公开实施例还提出一种非临时性计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如上述实施例所述的智能设备的交互目标确定方法。In order to implement the above embodiments, an embodiment of the present disclosure further provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by the processor, implements an interactive target of the smart device as described in the above embodiments. Determine the method.

在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本公开的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不必须针对的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。In the description of the present specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like means a specific feature described in connection with the embodiment or example. A structure, material, or feature is included in at least one embodiment or example of the present disclosure. In the present specification, the schematic representation of the above terms is not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, various embodiments or examples described in the specification, as well as features of various embodiments or examples, may be combined and combined.

此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。在本公开的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。Moreover, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, features defining "first" or "second" may include at least one of the features, either explicitly or implicitly. In the description of the present disclosure, the meaning of "a plurality" is at least two, such as two, three, etc., unless specifically defined otherwise.

流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现定制逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分，并且本公开的优选实施方式的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能，这应被本公开的实施例所属技术领域的技术人员所理解。Any process or method description in the flowcharts or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logic function or process. And the scope of the preferred embodiments of the present disclosure includes additional implementations, in which the functions may be performed in a substantially simultaneous manner or in an inverse order depending on the functions involved, in the order shown or discussed. It will be understood by those skilled in the art to which the embodiments of the present disclosure pertain.

在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行***、装置或设备(如基于计算机的***、包括处理器的***或其他可以从指令执行***、装置或设备取指令并执行指令的***)使用，或结合这些指令执行***、装置或设备而使用。就本说明书而言，"计算机可读介质"可以是任何可以包含、存储、通信、传播或传输程序以供指令执行***、装置或设备或结合这些指令执行***、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。The logic and/or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be embodied in any computer readable medium, Used in conjunction with, or in conjunction with, an instruction execution system, apparatus, or device (eg, a computer-based system, a system including a processor, or other system that can fetch instructions and execute instructions from an instruction execution system, apparatus, or device) Or use with equipment. For the purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport a program for use in an instruction execution system, apparatus, or device, or in conjunction with the instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer readable media include the following: electrical connections (electronic devices) having one or more wires, portable computer disk cartridges (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM). In addition, the computer readable medium may even be a paper or other suitable medium on which the program can be printed, as it may be optically scanned, for example by paper or other medium, followed by editing, interpretation or, if appropriate, other suitable The method is processed to obtain the program electronically and then stored in computer memory.

应当理解，本公开的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行***执行的软件或固件来实现。如，如果用硬件来实现和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。It should be understood that portions of the present disclosure can be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware and in another embodiment, it can be implemented by any one or combination of the following techniques well known in the art: discrete with logic gates for implementing logic functions on data signals Logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

本技术领域的普通技术人员可以理解实现上述实施例方法携带的全部或部分步骤是可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，该程序在执行时，包括方法实施例的步骤之一或其组合。One of ordinary skill in the art can understand that all or part of the steps carried by the method of implementing the above embodiments can be completed by a program to instruct related hardware, and the program can be stored in a computer readable storage medium. When executed, one or a combination of the steps of the method embodiments is included.

此外，在本公开各个实施例中的各功能单元可以集成在一个处理模块中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。In addition, each functional unit in various embodiments of the present disclosure may be integrated into one processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. The integrated modules, if implemented in the form of software functional modules and sold or used as stand-alone products, may also be stored in a computer readable storage medium.

上述提到的存储介质可以是只读存储器，磁盘或光盘等。尽管上面已经示出和描述了本公开的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本公开的限制，本领域的普通技术人员在本公开的范围内可以对上述实施例进行变化、修改、替换和变型。The above mentioned storage medium may be a read only memory, a magnetic disk or an optical disk or the like. While the embodiments of the present disclosure have been shown and described above, it is understood that the foregoing embodiments are illustrative and are not to be construed as limiting the scope of the disclosure The embodiments are subject to variations, modifications, substitutions and variations.

Claims

一种智能设备的交互目标确定方法，其特征在于，包括以下步骤：A method for determining an interaction target of a smart device, comprising the steps of:

获取在智能设备的监控范围内的环境图像，对所述环境图像进行目标识别；Obtaining an environment image within a monitoring range of the smart device, and performing target recognition on the environment image;

将从所述环境图像中识别出的目标作为候选目标，获取所述候选目标的状态信息；Obtaining the target information of the candidate target from the target identified in the environment image as a candidate target;

针对每个候选目标，根据对应的状态信息，判断是否存在与所述智能设备交互的交互意图；Determining, for each candidate target, whether there is an interaction intention of interacting with the smart device according to the corresponding state information;

从存在交互意图的候选目标中选取所述智能设备的交互目标。The interaction target of the smart device is selected from candidate objects with interaction intentions.
根据权利要求1所述的方法，其特征在于，所述获取所述候选目标的状态信息，包括：The method according to claim 1, wherein the obtaining the status information of the candidate target comprises:

获取所述候选目标与所述智能设备之间的距离；Obtaining a distance between the candidate target and the smart device;

针对每个候选目标，根据对应的状态信息，判断是否存在与所述智能设备交互的交互意图，包括：For each candidate target, determining, according to the corresponding state information, whether there is an interaction intention of interacting with the smart device, including:

针对每个候选目标，判断所述候选目标与所述智能设备之间的距离是否小于或者等于预设的所述距离阈值，且在所述距离阈值范围内的停留时长是否超出预设的时间阈值；Determining, for each candidate target, whether a distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether a stay duration within the distance threshold exceeds a preset time threshold ;

如果所述候选目标与所述智能设备之间的距离小于或者等于所述距离阈值且所述停留时长超出所述时间阈值，则确定所述该候选目标存在与所述智能设备交互的交互意图。If the distance between the candidate target and the smart device is less than or equal to the distance threshold and the stay duration exceeds the time threshold, determining that the candidate target has an interaction intention of interacting with the smart device.
根据权利要求1所述的方法，其特征在于，所述获取所述候选目标的状态信息，包括：The method according to claim 1, wherein the obtaining the status information of the candidate target comprises:

获取所述候选目标与所述智能设备之间的距离，以及所述候选目标的人脸角度；Obtaining a distance between the candidate target and the smart device, and a face angle of the candidate target;

针对每个候选目标，根据对应的状态信息，判断是否存在与智能设备交互的交互意图，包括：For each candidate target, according to the corresponding status information, it is determined whether there is an interaction intention of interacting with the smart device, including:

针对每个候选目标，判断所述候选目标与所述智能设备之间的距离是否小于等于预设的距离阈值，且所述候选目标的人脸角度是否处于预设的角度范围内；Determining, for each candidate target, whether the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether the face angle of the candidate target is within a preset angle range;

如果所述候选目标与智能设备之间的距离小于或等于预设的距离阈值，且所述候选目标的人脸角度处于预设的角度范围内，则确定所述候选目标存在与所述智能设备交互的交互意图。Determining that the candidate target exists with the smart device if the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and the face angle of the candidate target is within a preset angle range The interactive intent of the interaction.
根据权利要求1-3任一所述的方法，其特征在于，所述从存在交互意图的所述候选目标中选取所述智能设备的交互目标，包括：The method according to any one of claims 1-3, wherein the selecting an interaction target of the smart device from the candidate targets having an interaction intention comprises:

当检测到多个候选目标时，且存在交互意图的候选目标为多个时，从多个存在交互意图的所述候选目标中，确定出与所述智能设备距离最近的候选目标；When a plurality of candidate targets are detected, and there are a plurality of candidate targets having an interaction intention, determining, from the plurality of candidate targets having the interaction intention, a candidate target that is closest to the smart device;

从所述与所述智能设备距离最近的候选目标中，选取所述智能设备的交互目标。And selecting an interaction target of the smart device from the candidate target that is closest to the smart device.
根据权利要求4所述的方法，其特征在于，所述从所述与所述智能设备距离最近的候选目标中，选取所述智能设备的交互目标，包括：The method according to claim 4, wherein the selecting an interaction target of the smart device from the candidate target that is closest to the smart device comprises:

当与所述智能设备距离最近的候选目标为多个时，查询所述智能设备的已注册用户人脸图像库中是否存在与所述智能设备距离最近的候选目标的人脸图像；When there are a plurality of candidate targets that are closest to the smart device, query whether there is a face image of the candidate target that is closest to the smart device in the registered user face image library of the smart device;

如果所述人脸图像库中存在一个与所述智能设备距离最近的候选目标的人脸图像，则将所述一个与所述智能设备距离最近的候选目标作为交互目标；If there is a face image of a candidate target that is closest to the smart device in the face image library, the candidate target that is closest to the smart device is used as an interaction target;

如果所述人脸图像库中不存在与所述智能设备距离最近的候选目标的人脸图像，则随机选取一个与所述智能设备距离最近的候选目标作为交互目标；If there is no face image of the candidate target closest to the smart device in the face image library, randomly selecting a candidate target that is closest to the smart device as an interaction target;

如果所述人脸图像库中存在多个与所述智能设备距离最近的候选目标的人脸图像，则将最先查询出的与所述智能设备距离最近的候选目标作为交互目标。If there are a plurality of face images of the candidate target that are closest to the smart device in the face image library, the candidate target that is the closest to the smart device that is firstly queried is used as the interaction target.
根据权利要求2-5任一所述的方法，其特征在于，所述获取所述候选目标与所述智能设备人之间的距离，包括：The method according to any one of claims 2 to 5, wherein the obtaining the distance between the candidate target and the smart device person comprises:

通过所述智能设备中的深度摄像头获取深度图，根据所述深度图，获取所述目标与所述智能设备之间的距离；或者，Obtaining a depth map by using a depth camera in the smart device, and acquiring a distance between the target and the smart device according to the depth map; or

通过所述智能设备中的双目视觉摄像头，对所述候选目标进行拍摄，计算所述双目视觉摄像头所拍摄图像的视差，根据所述视差计算所述候选目标与所述智能设备之间的距离；或者，Obtaining, by the binocular vision camera in the smart device, the candidate target, calculating a disparity of the image captured by the binocular vision camera, and calculating, between the candidate target and the smart device, according to the disparity Distance; or,

通过所述智能设备中的激光雷达，向所述监控范围内发射激光；Laser light is emitted into the monitoring range by a laser radar in the smart device;

根据处于所述监控范围内的每个障碍物返回的激光，生成每个障碍物的二值图；Generating a binary map of each obstacle based on the laser light returned by each obstacle within the monitored range;

将每个二值图与所述环境图像进行融合，从所有的二值图中识别出与所述候选目标对应的二值图；Combining each binary image with the environment image, and identifying a binary image corresponding to the candidate target from all the binary images;

根据所述候选目标对应的二值图的激光返回时间，确定出所述候选目标与所述智能设备之间的距离。Determining a distance between the candidate target and the smart device according to a laser return time of the binary image corresponding to the candidate target.
根据权利要求3所述的方法，其特征在于，所述获取所述候选目标的人脸角度，包括：The method according to claim 3, wherein the obtaining a face angle of the candidate target comprises:

从所述环境图像中截取所述候选目标的人脸图像；Extracting a face image of the candidate target from the environment image;

将所述人脸图像输入预先训练好的机器学习模型中，获取所述人脸图像中人脸的人脸角度；Inputting the face image into a pre-trained machine learning model, and acquiring a face angle of the face in the face image;

所述方法还包括：采用如下方式训练所述机器学习模型：The method also includes training the machine learning model in the following manner:

采集样本人脸图像，其中，所述样本人脸图像中携带标注数据，所述标注数据用于表示样本人脸的人脸角度；Collecting a sample face image, wherein the sample face image carries annotation data, and the annotation data is used to represent a face angle of the sample face;

将所述样本人脸图像输入到初始构建的机器学习模型中进行训练，当训练后的所述机器学习模型的误差在预设的误差范围内时，则得到训练好的所述机器学习模型。The sample face image is input into the initially constructed machine learning model for training, and when the error of the trained machine learning model is within a preset error range, the trained machine learning model is obtained.
根据权利要求1-7任一所述的方法，其特征在于，所述从存在交互意图的候选目标中选取所述智能设备的交互目标之后，还包括：The method according to any one of claims 1 to 7, wherein after the selecting the interaction target of the smart device from the candidate objects having the interaction intention, the method further comprises:

控制所述智能设备与所述交互目标进行交互；Controlling the smart device to interact with the interaction target;

在交互过程中，识别所述交互目标的人脸图像的中心点；Identifying a center point of the face image of the interaction target during the interaction;

检测所述人脸图像的中心点是否在预设的图像区域内；Detecting whether a center point of the face image is within a preset image area;

如果未在所述图像区域内，获取所述人脸图像的中心点到达所述图像区域的中心点之间的路径；If not in the image area, acquiring a path between a center point of the face image and a center point of the image area;

根据所述路径，控制所述智能设备，使所述人脸图像的中心点在所述图像区域内。According to the path, the smart device is controlled such that a center point of the face image is within the image area.
一种智能设备的交互目标确定装置，其特征在于，包括：An interactive target determining apparatus for a smart device, comprising:

第一获取模块，用于获取在智能设备的监控范围内的环境图像，对所述环境图像进行目标识别；a first acquiring module, configured to acquire an environment image within a monitoring range of the smart device, and perform target recognition on the environment image;

第二获取模块，用于将从所述环境图像中识别出的目标作为候选目标，获取所述候选目标的状态信息；a second acquiring module, configured to acquire, as a candidate target, a target object that is identified from the environment image, and acquire state information of the candidate target;

判断模块，用于针对每个候选目标，根据对应的状态信息，判断是否存在与所述智能设备交互的交互意图；a determining module, configured to determine, according to the corresponding state information, whether there is an interaction intention of interacting with the smart device according to each candidate target;

选取模块，用于从存在交互意图的候选目标中选取所述智能设备的交互目标。And a selection module, configured to select an interaction target of the smart device from a candidate target having an interaction intention.
根据权利要求9所述的装置，其特征在于，所述第二获取模块具体用于：The device according to claim 9, wherein the second obtaining module is specifically configured to:

获取所述候选目标与所述智能设备之间的距离；Obtaining a distance between the candidate target and the smart device;

所述判断模块具体用于：The determining module is specifically configured to:

针对每个候选目标，判断所述候选目标与所述智能设备之间的距离是否小于或者等于预设的所述距离阈值，且在所述距离阈值范围内的停留时长是否超出预设的时间阈值；Determining, for each candidate target, whether a distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether a stay duration within the distance threshold exceeds a preset time threshold ;

如果所述距离小于或者等于所述距离阈值且所述停留时长超出所述时间阈值，则确定所述该候选目标存在与所述智能设备交互的交互意图。And if the distance is less than or equal to the distance threshold and the stay duration exceeds the time threshold, determining that the candidate target has an interaction intention of interacting with the smart device.
根据权利要求9所述的装置，其特征在于，所述第二获取模块具体用于：The device according to claim 9, wherein the second obtaining module is specifically configured to:

获取所述候选目标与所述智能设备之间的距离，以及所述候选目标的人脸角度；Obtaining a distance between the candidate target and the smart device, and a face angle of the candidate target;

所述判断模块具体用于：The determining module is specifically configured to:

针对每个候选目标，判断所述候选目标与所述智能设备之间的距离是否小于或者等于预设的距离阈值，且所述候选目标的人脸角度是否处于预设的角度范围内；Determining, for each candidate target, whether the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and whether the face angle of the candidate target is within a preset angle range;

如果所述候选目标与所述智能设备之间的距离小于或者等于预设的距离阈值，且所述候选目标的人脸角度处于预设的角度范围内，则确定所述候选目标存在与所述智能设备交互的交互意图。If the distance between the candidate target and the smart device is less than or equal to a preset distance threshold, and the face angle of the candidate target is within a preset angle range, determining that the candidate target exists and The interactive intent of the smart device interaction.
根据权利要求9-11任一所述的装置，其特征在于，所述选取模块包括：The device according to any one of claims 9-11, wherein the selection module comprises:

确定单元，用于当检测到多个候选目标时，且存在交互意图的候选目标为多个时，从多个存在交互意图的所述候选目标中，确定出与所述智能设备距离最近的候选目标；a determining unit, configured to: when a plurality of candidate targets are detected, and there are a plurality of candidate targets having an interaction intention, determine, from the plurality of candidate targets having the interaction intention, a candidate closest to the smart device aims;

选取单元，用于从所述与所述智能设备距离最近的候选目标中，选取所述智能设备的交互目标。And a selecting unit, configured to select an interaction target of the smart device from the candidate target that is closest to the smart device.
根据权利要求12所述的装置，其特征在于，所述选取单元具体用于：The device according to claim 12, wherein the selecting unit is specifically configured to:

当与所述智能设备距离最近的候选目标为多个时，查询所述智能设备的已注册用户人脸图像库中是否存在与所述智能设备距离最近的候选目标的人脸图像；When there are a plurality of candidate targets that are closest to the smart device, query whether there is a face image of the candidate target that is closest to the smart device in the registered user face image library of the smart device;

如果所述人脸图像库中存在一个与所述智能设备距离最近的候选目标的人脸图像，则将所述一个与所述智能设备距离最近的候选目标作为交互目标；If there is a face image of a candidate target that is closest to the smart device in the face image library, the candidate target that is closest to the smart device is used as an interaction target;

如果所述人脸图像库中不存在与所述智能设备距离最近的候选目标的人脸图像，则随机选取一个与所述智能设备距离最近的候选目标作为交互目标；If there is no face image of the candidate target closest to the smart device in the face image library, randomly selecting a candidate target that is closest to the smart device as an interaction target;

如果所述人脸图像库中存在多个与所述智能设备距离最近的候选目标的人脸图像，则将最先查询出的与所述智能设备距离最近的候选目标作为交互目标。If there are a plurality of face images of the candidate target that are closest to the smart device in the face image library, the candidate target that is the closest to the smart device that is firstly queried is used as the interaction target.
根据权利要求10-13任一所述的装置，其特征在于，所述第二获取模块具体用于：The device according to any one of claims 10-13, wherein the second obtaining module is specifically configured to:

通过所述智能设备中的深度摄像头获取深度图，根据所述深度图，获取所述目标与所述智能设备之间的距离；或者，Obtaining a depth map by using a depth camera in the smart device, and acquiring a distance between the target and the smart device according to the depth map; or

通过所述智能设备中的双目视觉摄像头，对所述候选目标进行拍摄，计算所述双目视觉摄像头所拍摄图像的视差，根据所述视差计算所述候选目标与所述智能设备之间的距离；或者，Obtaining, by the binocular vision camera in the smart device, the candidate target, calculating a disparity of the image captured by the binocular vision camera, and calculating, between the candidate target and the smart device, according to the disparity Distance; or,

通过所述智能设备中的激光雷达，向所述监控范围内发射激光；Laser light is emitted into the monitoring range by a laser radar in the smart device;

根据处于所述监控范围内的每个障碍物返回的激光，生成每个障碍物的二值图；Generating a binary map of each obstacle based on the laser light returned by each obstacle within the monitored range;

将每个二值图与所述环境图像进行融合，从所有的二值图中识别出与所述候选目标对应的二值图；Combining each binary image with the environment image, and identifying a binary image corresponding to the candidate target from all the binary images;

根据所述候选目标对应的二值图的激光返回时间，确定出所述候选目标与所述智能设备之间的距离。Determining a distance between the candidate target and the smart device according to a laser return time of the binary image corresponding to the candidate target.
根据权利要求11所述的装置，其特征在于，所述第二获取模块具体用于：The device according to claim 11, wherein the second obtaining module is specifically configured to:

从所述环境图像中截取所述候选目标的人脸图像；Extracting a face image of the candidate target from the environment image;

将所述人脸图像输入预先训练好的机器学习模型中，获取所述人脸图像中人脸的人脸角度；Inputting the face image into a pre-trained machine learning model, and acquiring a face angle of the face in the face image;

所述装置还包括：The device also includes:

采集模块，用于采集携带样本人脸图像，其中，所述样本人脸图像中携带标注数据，所述标注数据用于表示样本人脸的人脸角度；An acquisition module, configured to collect a face image of the sample, wherein the sample face image carries the annotation data, and the annotation data is used to represent a face angle of the sample face;

训练模块，用于将所述样本人脸图像输入到初始构建的机器学习模型中进行训练，当训练后的所述机器学习模型的误差在预设的误差范围内时，则得到训练好的所述机器学习模型。a training module, configured to input the sample face image into an initially constructed machine learning model for training, and when the error of the machine learning model after training is within a preset error range, obtain a trained training A machine learning model.
根据权利要求9-15任一所述的装置，其特征在于，还包括：The device according to any one of claims 9-15, further comprising:

第一控制模块，用于在所述从存在交互意图的候选目标中选取所述智能设备的交互目标之后，控制所述智能设备与所述交互目标进行交互；a first control module, configured to control the smart device to interact with the interaction target after selecting the interaction target of the smart device from the candidate target having the interaction intention;

识别模块，用于在交互过程中，识别所述交互目标的人脸图像的中心点；a recognition module, configured to identify a center point of the face image of the interaction target during the interaction process;

检测模块，用于检测所述人脸图像的中心点是否在预设的图像区域内；a detecting module, configured to detect whether a center point of the face image is within a preset image area;

第三获取模块，用于在未在所述图像区域内时，获取所述人脸图像的中心点到达所述图像区域的中心点之间的路径；a third acquiring module, configured to acquire a path between a center point of the face image and a center point of the image area when not in the image area;

第二控制模块，用于根据所述路径，控制所述智能设备，使所述人脸图像的中心点在所述图像区域内。And a second control module, configured to control the smart device according to the path, so that a center point of the face image is within the image area.
一种智能设备，其特征在于，包括：壳体、处理器、存储器、电路板和电源电路，其中，所述电路板安置在所述壳体围成的空间内部，所述处理器和所述存储器设置在所述电路板上；所述电源电路，用于为上述智能设备的各个电路或器件供电；所述存储器用于存储可执行程序代码；其中，所述处理器通过读取所述存储器中存储的可执行程序代码来运行与所述可执行程序代码对应的程序，以用于实现如权利要求1-8中任一所述的智能设备的交互目标确定方法。A smart device, comprising: a housing, a processor, a memory, a circuit board, and a power supply circuit, wherein the circuit board is disposed inside a space enclosed by the housing, the processor and the a memory is disposed on the circuit board; the power circuit is configured to supply power to each circuit or device of the smart device; the memory is configured to store executable program code; wherein the processor reads the memory by reading The executable program code stored therein runs a program corresponding to the executable program code for implementing the interactive object determining method of the smart device according to any one of claims 1-8.
一种非临时性计算机可读存储介质，其上存储有计算机程序，其特征在于，该程序被处理器执行时实现如权利要求1-8中任一所述的智能设备的交互目标确定方法。A non-transitory computer readable storage medium having stored thereon a computer program, wherein the program is executed by a processor to implement an interactive object determining method of the smart device according to any one of claims 1-8.