WO2020051959A1 - Image-based costume size measurement method and device - Google Patents

Abstract

An image-based costume size measurement method and device, relating to the field of costume design and resolving the problem in the prior art of low efficiency due to dependency on manual work, such that the measurement efficiency can be improved. The method comprises: receiving image information uploaded by a user, the image information being an image comprising a preset reference object and a costume to be measured and obtained by a camera at a preset photographing angle in a preset photographing environment; obtaining a pixel size and an actual size of the preset reference object and generating a ratio coefficient on the basis of the pixel size and the actual size; identifying the costume type of the costume to be measured; searching, according to the costume type, a key point identification model matching the costume type; invoking the key point identification model to identify key point feature information; computing first size information of the costume to be measured on the basis of the key point feature information, wherein the first size information is a pixel size; and computing second size information on the basis of the first size information and the ratio coefficient, wherein the second size information is an actual size.

Description

基于图片的服装尺寸测量方法及装置Picture-based clothing size measurement method and device

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

本申请要求于2018年09月10日提交中国专利局的申请号为2018110560582、名称为“基于图片的服装尺寸测量方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。This application claims priority from a Chinese patent application filed with the Chinese Patent Office on September 10, 2018 under the application number 2018110560582, entitled "Picture-based clothing size measurement method and device", the entire contents of which are incorporated herein by reference. in.

技术领域Technical field

本申请涉及服装设计技术领域，尤其涉及基于图片的服装尺寸测量方法及装置。The present application relates to the technical field of clothing design, and in particular, to a method and a device for measuring the size of clothing based on pictures.

背景技术Background technique

服装是随着人类社会的发展而不断发展变化的。在远古时期，人类最初的衣服是用树叶、兽皮等制成的，之后，出现了用麻类纤维和草制等成的衣物。而随着生产力的提高和科技的发展，材料变得越来越丰富多样，相应的，制造的衣物的种类也逐渐增多。在现代社会中，人们可以根据自己的喜好和具体需求来选择适合自己的服装，从一定意义上来说，服装已经是每个人装饰自己，保护自己，能给自己和家人的必用品，不仅仅为穿，更是一个身份、一种生活态度、一个展示个人魅力的表现。Clothing is constantly changing with the development of human society. In ancient times, the first clothes of humans were made of leaves, animal skins, etc. After that, clothes made of hemp fiber and grass appeared. With the improvement of productivity and the development of technology, materials have become more and more diverse, and accordingly, the types of clothing manufactured have gradually increased. In modern society, people can choose clothing that suits them according to their preferences and specific needs. In a certain sense, clothing is already an essential item for everyone to decorate themselves, protect themselves, and give themselves and their families. Wearing is more an identity, an attitude towards life, and a manifestation of personal charm.

因此，在服装设计的过程中需要对服装进行测量，以得到服装的测量值尺寸，然后基于该尺寸测量值进行后续改进以适合更多消费者的需求。然而，传统的服装测量方法依靠人工测量的方式进行，存在效率低下的问题。Therefore, in the process of clothing design, it is necessary to measure the clothing to obtain the measured value size of the clothing, and then perform subsequent improvements based on the size measurement value to meet the needs of more consumers. However, traditional clothing measurement methods rely on manual measurement, which has the problem of low efficiency.

发明内容Summary of the Invention

有鉴于此，本申请实施例的目的在于提供了基于图片的服装尺寸测量方法及装置，一定程度上可以缓解现有技术中依靠人工的方式存在的效率低下的问题，提高测量效率。In view of this, the purpose of the embodiments of the present application is to provide a picture-based clothing size measurement method and device, which can alleviate to a certain extent the problem of inefficiency in the prior art relying on manual methods and improve measurement efficiency.

第一方面，本申请实施例提供了基于图片的服装尺寸测量方法，包括：In a first aspect, an embodiment of the present application provides a picture-based clothing size measurement method, including:

接收用户上传的图片信息，所述图片信息为包括预设参照物和待测量服装的图片；所述图片信息是在预设拍照环境、由摄像头在预设拍照角度下获得的；Receiving picture information uploaded by a user, the picture information being a picture including a preset reference object and clothing to be measured; the picture information is obtained in a preset photographing environment and a camera at a preset photographing angle;

获取所述图片信息中的所述预设参照物的像素尺寸数据以及实际尺寸数据，基于所述像素尺寸数据和所述实际尺寸数据生成比率系数；Acquiring pixel size data and actual size data of the preset reference object in the picture information, and generating a ratio coefficient based on the pixel size data and the actual size data;

识别所述图片信息中的所述待测量服装的服装类别；Identifying a clothing category of the clothing to be measured in the picture information;

根据所述服装类别查找与所述服装类别相匹配的服装关键点识别模型；Find a clothing keypoint recognition model matching the clothing category according to the clothing category;

调用所述服装关键点识别模型识别所述图片信息中的服装关键点特征信息；Calling the clothing keypoint recognition model to identify clothing keypoint feature information in the picture information;

基于所述服装关键点特征信息进行计算得到所述待测量服装的第一尺寸信息；其中所述第一尺寸信息包括所述待测量服装的预设特征部位的像素尺寸；Calculate the first size information of the clothing to be measured based on the clothing keypoint feature information; wherein the first size information includes a pixel size of a preset feature part of the clothing to be measured;

基于所述第一尺寸信息和所述比率系数计算得到第二尺寸信息；其中所述第二尺寸信 息包括所述待测量服装的所述预设特征部位的实际尺寸。A second size information is calculated based on the first size information and the ratio coefficient; wherein the second size information includes an actual size of the preset feature part of the clothing to be measured.

结合第一方面，本申请实施例提供了第一方面的第一种可能的实施方式，其中，所述调用所述服装关键点识别模型识别所述图片中的服装关键点特征信息，包括：With reference to the first aspect, the embodiment of the present application provides a first possible implementation manner of the first aspect, wherein the invoking the clothing keypoint recognition model to identify the clothing keypoint feature information in the picture includes:

调用所述服装关键点识别模型识别所述图片中的服装关键点类别信息及所述服装关键点位置信息。Calling the clothing keypoint recognition model to identify clothing keypoint category information and the clothing keypoint position information in the picture.

结合第一方面，本申请实施例提供了第一方面的第二种可能的实施方式，其中，所述图片信息中包括平铺状态下的待测量服装的图片。With reference to the first aspect, the embodiment of the present application provides a second possible implementation manner of the first aspect, wherein the picture information includes a picture of the clothing to be measured in a tiled state.

结合第一方面，本申请实施例提供了第一方面的第三种可能的实施方式，其中，所述预设参照物为预设形状的卡纸；所述卡纸上包括有服装识别码，所述服装识别码与所述电子设备所述服装识别码仅能够被所述电子设备识别；所述电子设备仅能够识别所述服装识别码。With reference to the first aspect, the embodiment of the present application provides a third possible implementation manner of the first aspect, wherein the preset reference object is a preset shape cardboard; the cardboard includes a clothing identification code, The clothing identification code and the electronic device can only be recognized by the electronic device; the electronic device can only identify the clothing identification code.

结合第一方面，本申请实施例提供了第一方面的第四种可能的实施方式，其中，所述预设拍照环境是通过摄影棚控制调节得到。With reference to the first aspect, the embodiment of the present application provides a fourth possible implementation manner of the first aspect, wherein the preset photographing environment is obtained through studio control adjustment.

结合第一方面，本申请实施例提供了第一方面的第五种可能的实施方式，其中，所述预设拍照角度是通过摄影架控制调节得到，以减少摄像头水平拍照时角度畸变；所述摄影架构造成设置所述摄像头。With reference to the first aspect, the embodiment of the present application provides a fifth possible implementation manner of the first aspect, in which the preset photographing angle is obtained through control of a photographic frame to reduce angle distortion when the camera is horizontally photographing; The photographic architecture results in setting up the camera.

结合第一方面，本申请实施例提供了第一方面的第六种可能的实施方式，其中，所述方法还包括：With reference to the first aspect, the embodiment of the present application provides a sixth possible implementation manner of the first aspect, where the method further includes:

将所述第二尺寸信息输出。And outputting the second size information.

结合第一方面，本申请实施例提供了第一方面的第七种可能的实施方式，其中，所述基于所述服装关键点特征信息进行计算得到所述待测量服装的第一尺寸信息，包括：With reference to the first aspect, the embodiment of the present application provides a seventh possible implementation manner of the first aspect, wherein the first size information of the clothing to be measured is obtained by calculating based on the clothing keypoint feature information, including: :

调取与所述服装类别相对应的关键点计算规则；其中所述关键点计算规则包括所述服装类别的预设特征部位的像素尺寸计算规则；Calling a key point calculation rule corresponding to the clothing category; wherein the key point calculation rule includes a pixel size calculation rule of a preset feature part of the clothing category;

基于所述服装关键点特征信息以及所述关键点计算规则进行计算得到所述第一尺寸信息。Calculate and obtain the first size information based on the clothing keypoint feature information and the keypoint calculation rule.

结合第一方面，本申请实施例提供了第一方面的第八种可能的实施方式，其中，所述方法还包括：With reference to the first aspect, the embodiment of the present application provides an eighth possible implementation manner of the first aspect, where the method further includes:

获取多个第一训练用服装图片，其中，所述第一训练用服装图片为标注有服装类别的图片；Obtaining a plurality of first training clothing pictures, wherein the first training clothing pictures are pictures marked with a clothing category;

基于所述第一训练用服装图片对卷积神经网络模型进行机器学习训练，以构建得到所述服装类别识别模型。Machine learning training is performed on the convolutional neural network model based on the first training clothing picture to construct and obtain the clothing category recognition model.

结合第一方面，本申请实施例提供了第一方面的第九种可能的实施方式，其中，所述 方法还包括：With reference to the first aspect, the embodiment of the present application provides a ninth possible implementation manner of the first aspect, where the method further includes:

获取多个第二训练用服装图片，所述第二训练用服装图片为标注有服装关键点特征信息的图片；Obtaining a plurality of second training clothing pictures, where the second training clothing pictures are pictures labeled with key point feature information of the clothing;

基于所述第二训练用服装图片对卷积神经网络模型进行机器学习训练，以构建得到所述服装关键点识别模型。Machine learning training is performed on the convolutional neural network model based on the second training clothing picture to construct the clothing keypoint recognition model.

第二方面，本申请实施例提供了基于图片的服装尺寸测量装置，包括：In a second aspect, an embodiment of the present application provides a picture-based clothing size measuring device, including:

图片接收模块，配置成接收用户上传的图片信息，所述图片信息为包括预设参照物和待测量服装的图片；所述图片信息是在预设拍照环境、由摄像头在预设拍照角度下获得的；A picture receiving module configured to receive picture information uploaded by a user, the picture information being a picture including a preset reference object and clothing to be measured; the picture information is obtained in a preset photographing environment and by a camera at a preset photographing angle of;

比例获取模块，配置成获取所述图片信息中的所述预设参照物的像素尺寸数据以及实际尺寸数据，基于所述像素尺寸数据和所述实际尺寸数据生成比率系数；A scale obtaining module configured to obtain pixel size data and actual size data of the preset reference object in the picture information, and generate a ratio coefficient based on the pixel size data and the actual size data;

类别识别模块，配置成识别所述图片信息中的所述待测量服装的服装类别；A category identification module configured to identify a clothing category of the clothing to be measured in the picture information;

查找模块，配置成根据所述服装类别查找与所述服装类别相匹配的服装关键点识别模型；A search module configured to find a clothing keypoint recognition model matching the clothing category according to the clothing category;

关键点识别模块，配置成调用所述服装关键点识别模型识别所述图片信息中的服装关键点特征信息；A key point recognition module configured to call the clothing key point recognition model to identify clothing key point feature information in the picture information;

第一计算模块，配置成基于所述服装关键点特征信息进行计算得到所述待测量服装的第一尺寸信息；其中所述第一尺寸信息包括所述待测量服装的预设特征部位的像素尺寸；A first calculation module configured to perform calculation based on the clothing keypoint feature information to obtain first size information of the clothing to be measured; wherein the first size information includes a pixel size of a preset feature portion of the clothing to be measured ;

第二计算模块，配置成基于所述第一尺寸信息和所述比率系数计算得到第二尺寸信息；其中所述第二尺寸信息包括所述待测量服装的所述预设特征部位的实际尺寸。A second calculation module configured to calculate and obtain second size information based on the first size information and the ratio coefficient; wherein the second size information includes an actual size of the preset feature part of the clothing to be measured.

第三方面，本申请实施例还提供一种电子设备，包括存储器以及处理器，存储器配置成存储支持处理器执行上述方面提供的基于图片的服装尺寸测量方法的程序，处理器配置成行存储器中存储的程序。According to a third aspect, an embodiment of the present application further provides an electronic device including a memory and a processor. The memory is configured to store a program that supports the processor to execute the picture-based clothing size measurement method provided in the above aspect, and the processor is configured to store in a line memory. program of.

第四方面，本申请实施例还提供一种计算机可读存储介质，计算机可读存储介质上存储有计算机程序，计算机程序被处理器运行时执行上述任一项的方法的步骤。According to a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is run by a processor, executes the steps of any of the foregoing methods.

本申请实施例提供的基于图片的服装尺寸测量方法、装置、电子设备及计算机可读存储介质，其中，该基于图片的服装尺寸测量方法包括：接收用户上传的图片信息，图片信息为包括预设参照物和待测量服装的图片；图片信息是在预设拍照环境、由摄像头在预设拍照角度下获得的；获取图片信息中的预设参照物的像素尺寸数据以及实际尺寸数据，基于像素尺寸数据和实际尺寸数据生成比率系数；识别图片信息中的待测量服装的服装类别；根据服装类别查找与服装类别相匹配的服装关键点识别模型；调用服装关键点识别模型识别图片信息中的服装关键点特征信息；基于服装关键点特征信息进行计算得到待测量服装的第一尺寸信息；其中第一尺寸信息包括待测量服装的预设特征部位的像素尺寸；基于第 一尺寸信息和比率系数计算得到第二尺寸信息；其中第二尺寸信息包括待测量服装的预设特征部位的实际尺寸。因此，本申请实施例提供的技术方案，缓解了现有技术中依靠人工的方式存在的效率低下的问题，能够提高测量效率。该方法基于用户输入的图片，对图片进行测量得到服装的多个部位的尺寸测量值，相比于现有技术中依靠人工的方式存在的效率低下的问题，能够提高测量效率，该方法简单、方便、快速，大大提高了服装尺寸的测量效率，节省了人力、时间成本。The image-based clothing size measurement method, device, electronic device, and computer-readable storage medium provided in the embodiments of the present application, wherein the image-based clothing size measurement method includes: receiving picture information uploaded by a user, and the picture information includes a preset Pictures of the reference object and the clothing to be measured; the picture information is obtained in the preset photographing environment and by the camera at the preset photographing angle; the pixel size data and actual size data of the preset reference object in the picture information are obtained, based on the pixel size Data and actual size data to generate a ratio coefficient; identify the clothing category of the clothing to be measured in the picture information; find the clothing keypoint recognition model that matches the clothing category according to the clothing category; call the clothing keypoint recognition model to identify the clothing key in the picture information Point feature information; calculation based on the key point feature information of the clothing to obtain the first size information of the clothing to be measured; wherein the first size information includes the pixel size of the predetermined feature part of the clothing to be measured; calculated based on the first size information and the ratio coefficient Second size information; The second size information includes the actual size of the predetermined characteristic part of the clothing to be measured. Therefore, the technical solutions provided in the embodiments of the present application alleviate the problem of low efficiency in the prior art relying on manual methods, and can improve measurement efficiency. This method is based on a picture input by a user, and measures the pictures to obtain the size measurement values of multiple parts of the garment. Compared with the problem of inefficiency of relying on manual methods in the prior art, the method can improve measurement efficiency. Convenient and fast, greatly improving the measurement efficiency of clothing size, saving labor and time costs.

本申请的其他特征和优点将在随后的说明书中阐述，并且，部分地从说明书中变得显而易见，或者通过实施本申请而了解。本申请的目的和其他优点在说明书、权利要求书以及附图中所特别指出的结构来实现和获得。Other features and advantages of the present application will be explained in the following description, and partly become apparent from the description, or be understood by implementing the present application. The objectives and other advantages of the present application are achieved and obtained in the structures particularly pointed out in the description, the claims, and the drawings.

为使本申请的上述目的、特征和优点能更明显易懂，下文特举较佳实施例，并配合所附附图，作详细说明如下。In order to make the above-mentioned objects, features, and advantages of this application more comprehensible, preferred embodiments are described below in conjunction with the accompanying drawings and described in detail below.

附图说明BRIEF DESCRIPTION OF THE DRAWINGS

为了更清楚地说明本申请具体实施方式或现有技术中的技术方案，下面将对具体实施方式或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施方式，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the specific implementations of the present application or the technical solutions in the prior art, the drawings used in the specific implementations or prior art descriptions will be briefly introduced below. Obviously, the appended The drawings are some implementations of the present application. For those of ordinary skill in the art, other drawings can be obtained according to these drawings without paying creative labor.

图1示出了本申请实施例所提供的电子设备的示意图；FIG. 1 is a schematic diagram of an electronic device according to an embodiment of the present application; FIG.

图2示出了本申请实施例所提供的基于图片的服装尺寸测量方法的流程图；FIG. 2 shows a flowchart of a picture-based clothing size measurement method according to an embodiment of the present application;

图3示出了本申请实施例所提供的一种图片信息的示意图；FIG. 3 is a schematic diagram of picture information provided by an embodiment of the present application; FIG.

图4示出了本申请实施例所提供的一种服装图片中的关键点特征信息的示意图；FIG. 4 is a schematic diagram of key point feature information in a clothing picture according to an embodiment of the present application; FIG.

图5示出了本申请实施例所提供的基于图片的服装尺寸测量方法的应用场景图；FIG. 5 illustrates an application scenario diagram of a picture-based clothing size measurement method according to an embodiment of the present application; FIG.

图6示出了本申请实施例所提供的基于图片的服装尺寸测量装置的结构示意图。FIG. 6 is a schematic structural diagram of a picture-based clothing size measuring device according to an embodiment of the present application.

具体实施方式detailed description

下面将结合本申请实施例中附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。通常在此处附图中描述和示出的本申请实施例的组件可以以各种不同的配置来布置和设计。因此，以下对在附图中提供的本申请的实施例的详细描述并非旨在限制要求保护的本申请的范围，而是仅仅表示本申请的选定实施例。基于本申请的实施例，本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都属于本申请保护的范围。In the following, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all of the embodiments. The components of embodiments of the present application, which are generally described and illustrated in the drawings herein, may be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work fall into the protection scope of the present application.

目前，在服装设计的过程中需要对服装进行测量，以得到服装的测量值尺寸，然后基于该尺寸测量值进行后续改进以适合更多消费者的需求。然而，传统的服装测量一种是依靠人工测量的方式进行，工作效率低；另一种是将衣服套在立体人形模型上进行立体人形 模型拍照，拍照效率低；基于此，本申请实施例提供了基于图片的服装尺寸测量方法及装置，以缓解现有技术中依靠人工存在的效率低下的问题，能够提高测量效率。At present, in the process of clothing design, it is necessary to measure the clothing to obtain the measured value size of the clothing, and then perform subsequent improvements based on the size measurement value to meet the needs of more consumers. However, the traditional clothing measurement is performed by manual measurement, which has low work efficiency; the other is that the clothes are put on the three-dimensional humanoid model to take pictures of the three-dimensional humanoid model, which has low efficiency; based on this, the embodiments of the present application provide A picture-based clothing size measurement method and device are provided to alleviate the problem of inefficiency due to manual labor in the prior art and improve measurement efficiency.

下面通过实施例进行详细描述。Detailed descriptions are provided below through examples.

实施例一：Embodiment one:

参见图1，本申请实施例还提供一种电子设备100，包括：处理器40，存储器41，总线42和通信接口43，上述处理器40、通信接口43和存储器41通过总线42连接；处理器40配置成执行存储器41中存储的可执行模块，例如计算机程序。Referring to FIG. 1, an embodiment of the present application further provides an electronic device 100 including: a processor 40, a memory 41, a bus 42, and a communication interface 43. The processor 40, the communication interface 43, and the memory 41 are connected through the bus 42; the processor 40 is configured to execute an executable module stored in the memory 41, such as a computer program.

其中，存储器41可能包含高速随机存取存储器(RAM，Random Access Memory)，也可能还包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。通过至少一个通信接口43(可以是有线或者无线)实现该***网元与至少一个其他网元之间的通信连接，可以使用互联网，广域网，本地网，城域网等。The memory 41 may include a high-speed random access memory (RAM, Random Access Memory), and may also include a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. The communication connection between the network element of the system and at least one other network element is realized through at least one communication interface 43 (which can be wired or wireless), and the Internet, a wide area network, a local network, a metropolitan area network, and the like can be used.

总线42可以是ISA总线、PCI总线或EISA总线等。上述总线可以分为地址总线、数据总线、控制总线等。为便于表示，图6中仅用一个双向箭头表示，但并不表示仅有一根总线或一种类型的总线。The bus 42 may be an ISA bus, a PCI bus, an EISA bus, or the like. The above buses can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only a two-way arrow is used in FIG. 6, but it does not mean that there is only one bus or one type of bus.

其中，存储器41配置成存储程序，上述处理器40在接收到执行指令后，执行上述程序，下述本申请实施例任一实施例揭示的流过程定义的***所执行的方法可以应用于处理器40中，或者由处理器40实现。The memory 41 is configured to store a program, and the processor 40 executes the program after receiving the execution instruction. The method executed by the system defined by the flow process disclosed in any one of the embodiments of the present application described below may be applied to the processor. 40, or implemented by the processor 40.

处理器40可能是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法的各步骤可以通过处理器40中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器40可以是通用处理器，包括中央处理器(Central Processing Unit，简称CPU)、网络处理器(Network Processor，简称NP)等；还可以是数字信号处理器(Digital Signal Processing，简称DSP)、专用集成电路(Application Specific Integrated Circuit，简称ASIC)、现成可编程门阵列(Field-Programmable Gate Array，简称FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器41，处理器40读取存储器41中的信息，结合其硬件完成上述方法的步骤。The processor 40 may be an integrated circuit chip and has a signal processing capability. In the implementation process, each step of the above method may be completed by an integrated logic circuit of hardware in the processor 40 or an instruction in the form of software. The above-mentioned processor 40 may be a general-purpose processor, including a central processing unit (CPU), a network processor (Network Processor) (NP), etc .; it may also be a digital signal processor (Digital Signal Processing, DSP) ), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. Various methods, steps, and logical block diagrams disclosed in the embodiments of the present application can be implemented or executed. A general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in combination with the embodiments of the present application may be directly implemented by a hardware decoding processor, or may be performed by using a combination of hardware and software modules in the decoding processor. The software module may be located in a mature storage medium such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, and the like. The storage medium is located in the memory 41, and the processor 40 reads the information in the memory 41 and completes the steps of the foregoing method in combination with its hardware.

本申请实施例还提供一种计算机可读存储介质，计算机可读存储介质上存储有计算机 程序，计算机程序被处理器运行时执行下述任一项的方法的步骤。An embodiment of the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is run by a processor, executes the steps of any one of the methods described below.

本申请实施例提供了一种基于图片的服装尺寸测量方法，可应用于服装设计领域，尤其适用于服装的尺寸测量。该方法由电子设备执行，上述电子设备包括API(Application Programming Interface，应用程序编程接口)，电子设备通过该API可以接收用户上传的图片信息。The embodiment of the present application provides a picture-based clothing size measurement method, which can be applied to the field of clothing design, and is particularly suitable for clothing size measurement. The method is executed by an electronic device. The electronic device includes an API (Application Programming Interface), and the electronic device can receive picture information uploaded by a user through the API.

具体的，参照图2，该方法包括：Specifically, referring to FIG. 2, the method includes:

步骤S101，接收用户上传的图片信息，该图片信息为包括预设参照物和待测量服装的图片。Step S101: Receive picture information uploaded by a user, where the picture information is a picture including a preset reference object and clothing to be measured.

在本实施例中，电子设备100可以通过通信接口43接收用户上传的图片信息。In this embodiment, the electronic device 100 may receive the picture information uploaded by the user through the communication interface 43.

这里需要说明的是，所述图片信息是在预设拍照环境、由摄像头在预设拍照角度下获得的；其中，所述预设拍照环境是通过摄影棚控制调节得到，以减少噪声物体对图片识别的影响。It should be noted here that the picture information is obtained in a preset photographing environment and a camera at a preset photographing angle; wherein the preset photographing environment is obtained through studio control adjustment to reduce noise objects to the picture Identify the impact.

通过用摄影棚控制拍摄环境得到预设拍照环境(即适于拍摄出清晰的图片的光强等良好的拍照环境)，可以减少杂音物体影响AI(Artificial Intelligence，人工智能)视觉算法这里的AI包括服装识别AI(服装类别识别模型)和服装关键点AI(服装关键点识别模型)。By using the studio to control the shooting environment to obtain a preset shooting environment (that is, a good shooting environment suitable for taking clear pictures of light intensity), it can reduce the influence of noise objects on the AI (Artificial Intelligence) vision algorithm. The AI here includes Clothing recognition AI (clothing category recognition model) and clothing keypoint AI (clothing keypoint recognition model).

上述的预设拍照角度是通过摄影架控制调节得到，以减少摄像头水平拍照时角度畸变；所述摄影架构造成设置所述摄像头；所述摄影架设置在上述的摄像棚中。也就是说，通过采用位于摄像棚的摄影架控制设置在摄影架上的摄像头水平，减少拍照角度畸变。The above-mentioned preset photographing angle is obtained by adjusting and adjusting the photographic frame to reduce the angle distortion when the camera is horizontally photographed; the photographic framework causes the camera to be set; and the photographic frame is disposed in the aforementioned video studio. That is to say, by using a photography rack located in the camera studio to control the level of the camera set on the photography rack, the distortion of the photographing angle is reduced.

相比与现有技术中的将衣服套在立体人形模型上进行立体人形模型拍照的方式，本申请实施例采用将待测量服装平铺拍照的方式，提高了拍照效率。Compared with the conventional method of taking clothes on a three-dimensional humanoid model and taking pictures of the three-dimensional humanoid model, the embodiment of the present application adopts a method of taking photos of the clothing to be measured in a tiled manner, thereby improving the efficiency of taking pictures.

进一步的，所述预设参照物为预设形状的卡纸，所述卡纸上包括有服装识别码，所述服装识别码仅能够被所述服装尺寸识别***识别。所述服装尺寸识别***仅能够识别所述服装识别码。上述的服装识别码是一种AR(Augmented Reality，增强现实)代码，可以通过AR编码技术预先编码得到。Further, the preset reference object is a cardboard with a preset shape, and the cardboard includes a clothing identification code, and the clothing identification code can only be recognized by the clothing size recognition system. The clothing size recognition system can only recognize the clothing identification code. The above-mentioned clothing identification code is an AR (Augmented Reality, Augmented Reality, AR) code, which can be obtained by encoding in advance through an AR coding technology.

图3示出了一种图片信息的示意图。具体的，参见图3，所述预设参照物为打印有服装识别码的特定尺寸的矩形卡纸。也就是说，矩形卡纸的长度、宽度的尺寸是已知的；当然，预设参照物还可以采用平面型的其他任意形状。FIG. 3 shows a schematic diagram of picture information. Specifically, referring to FIG. 3, the preset reference object is a rectangular cardboard of a specific size printed with a clothing identification code. In other words, the dimensions of the length and width of the rectangular cardboard are known; of course, the preset reference object can also adopt any other shape of a flat type.

本申请实施例通过利用AR编码技术生成固定编码(服装识别码)并且将其打印在特定尺寸卡纸上，卡纸作为平面形的预设参照物，与平放的待测量服装一起拍照，由于电子设备仅能识别已设置的固定编码，防止了其他形式的码型(例如二维码、条形码等)对电子设备的影响，通过排除干扰因素，确保了服装识别的唯一性，提高了服装识别的准确性 和高效性。In the embodiment of the present application, a fixed code (clothing identification code) is generated by using the AR coding technology and printed on a specific size cardboard, and the cardboard is used as a flat preset reference, and is photographed together with the clothing to be measured in a flat position. The electronic device can only recognize the fixed code that has been set, preventing the impact of other forms of codes (such as two-dimensional codes, barcodes, etc.) on the electronic device. By eliminating interference factors, it ensures the uniqueness of clothing recognition and improves clothing recognition. Accuracy and efficiency.

步骤S102，获取图片信息中的预设参照物的像素尺寸数据和实际尺寸数据，基于像素尺寸数据和实际尺寸数据生成比率系数。Step S102: Obtain pixel size data and actual size data of a preset reference object in the picture information, and generate a ratio coefficient based on the pixel size data and the actual size data.

在本实施例中，可以由电子设备100的处理器40从图片信息中获取预设参照物的像素尺寸数据和实际尺寸数据。In this embodiment, the processor 40 of the electronic device 100 may obtain pixel size data and actual size data of a preset reference object from the picture information.

这里的像素尺寸数据是指预设参照物的预设位置的像素尺寸，像素尺寸数据可以直接通过图片读取得到。这里的实际尺寸数据是指预设参照物的预设位置的实际尺寸；上述的预设参照物的实际尺寸数据是已知的，且可以预先存储在电子设备的存储器中以方便调用，实际尺寸数据通过从存储器中调取(提取)得到。The pixel size data here refers to the pixel size of the preset position of the preset reference object, and the pixel size data can be directly obtained by reading the picture. The actual size data here refers to the actual size of the preset position of the preset reference object; the above-mentioned actual size data of the preset reference object is known and can be stored in advance in the memory of the electronic device for convenient recall, the actual size The data is obtained by fetching (extracting) from the memory.

例如预设位置可以是预设参照物的轮廓的一条边线，该预设参照物的轮廓的一条边线可以是任意两个轮廓边缘的像素点连接而成的直线。For example, the preset position may be an edge of the outline of the preset reference object, and an edge of the outline of the preset reference object may be a straight line formed by connecting pixels of any two contour edges.

为了计算比率系数，可以将该预设参照物的预设位置(例如上述预设参照物的轮廓的一条边线)的实际尺寸与图片信息中的相同的该预设位置的像素尺寸之比即得到比率系数。In order to calculate the ratio coefficient, the ratio between the actual size of the preset position of the preset reference object (such as an edge of the outline of the preset reference object) and the pixel size of the same preset position in the picture information is obtained. Ratio factor.

具体的，可以利用识别算法计算预设参照物的边线的像素距离(即像素尺寸数据)，再参考已设置好的预设参照物的该边线的实际尺寸(即实际尺寸数据)，用实际尺寸除以像素距离计算出比率系数，这里实际尺寸可以用毫米来度量，计算出的比率系数为毫米对像素比率。这里的比率系数用于表征实际尺寸数据和像素尺寸的换算关系，实际尺寸数据的单位并无影响，因此实际尺寸数据也可以以厘米等其他长度单位来度量，这里不作限定。Specifically, the recognition algorithm can be used to calculate the pixel distance (ie, pixel size data) of the edge of the preset reference object, and then refer to the actual size (ie, actual size data) of the edge of the preset reference object that has been set, and use the actual size Divide the pixel distance to calculate the ratio coefficient. Here the actual size can be measured in millimeters. The calculated ratio coefficient is the millimeter-to-pixel ratio. The ratio coefficient here is used to characterize the conversion relationship between the actual size data and the pixel size. The unit of the actual size data has no effect, so the actual size data can also be measured in other length units such as centimeters, which is not limited here.

为了便于理解，这里以图2中的矩形卡纸作为预设参照物，预设位置为矩形的长边(即矩形的长)为例对上述比率系数的计算过程进行说明：For ease of understanding, the rectangular jammed paper in Figure 2 is used as a preset reference, and the preset position is the long side of the rectangle (that is, the length of the rectangle) as an example to explain the calculation process of the above-mentioned ratio coefficient:

首先，获取图2中矩形卡纸的长边的像素尺寸以及从存储器调取已知的该矩形卡纸的长边的实际尺寸；然后基于两者进行计算得到比率系数，具体的，将该长边的实际尺寸除以该长边的像素尺寸即得到比率系数。First, obtain the pixel size of the long side of the rectangular cardboard in FIG. 2 and retrieve the known actual size of the long side of the rectangular cardboard from the memory; then calculate the ratio coefficient based on the two. Specifically, the length The actual size of the side divided by the pixel size of the long side gives the ratio coefficient.

步骤S103，识别图片信息中的待测量服装的服装类别。Step S103: Identify the clothing category of the clothing to be measured in the picture information.

在本实施例中，可以由电子设备100的处理器40识别图片信息中的待测量服装的服装类别。In this embodiment, the processor 40 of the electronic device 100 can identify the clothing category of the clothing to be measured in the picture information.

具体的，处理器40调取预先构建的服装类别识别模型识别该图片信息中的待测量服装的服装类别；上述的服装类别包括T恤、裤子、衬衫、裙子等服装款式。Specifically, the processor 40 calls a pre-built clothing category recognition model to identify the clothing category of the clothing to be measured in the picture information; the aforementioned clothing categories include clothing styles such as T-shirts, pants, shirts, and skirts.

预先构建的服装类别识别模型可以存储在电子设备100的存储器41中，以便于后续调用及服装类别识别。The pre-built clothing category recognition model may be stored in the memory 41 of the electronic device 100 to facilitate subsequent calls and clothing category recognition.

因此，该方法还包括：构建服装类别识别模型，该构建服装类别识别模型通过以下步骤执行：Therefore, the method further includes: constructing a clothing category recognition model, and the constructing a clothing category recognition model is performed by the following steps:

1、获取第一训练用服装图片；其中第一训练用服装图片为标注有服装类别的图片；第一训练用服装为多件，且，多件所述第一训练用服装对应的服装类别均不同。1. Obtain a picture of the first training clothing; wherein the picture of the first training clothing is a picture labeled with a clothing category; there are multiple pieces of the first training clothing, and the clothing categories corresponding to the multiple pieces of the first training clothing are all different.

第一训练用服装图片是在上述预设拍照环境、由摄像头在预设拍照角度下获得的，即摄像头分别对多件训练用服装进行拍摄生成图片，然后在该图片上进行服装类别标注，得到训练用服装图片；The first training clothing picture is obtained in the preset photographing environment and the camera at a preset photographing angle, that is, the camera takes pictures of multiple training clothings to generate pictures, and then labels the clothing category on the picture to obtain Training clothing pictures;

2、基于上述第一训练用服装图片对CNN(Convolutional Neural Network，卷积神经网络)模型进行训练，以构建得到服装类别识别模型。2. Train a CNN (Convolutional Neural Network, Convolutional Neural Network) model based on the first training clothing picture to construct a clothing category recognition model.

具体的，基于多张所述第一训练用图片通过监督式算法对CNN模型进行训练，构建服装类别识别模型。即通过深度学习的监督式算法训练CNN模型对第一训练用服装图片中的服装类别进行学习和识别，构建服装类别识别模型，构建的服装类别识别模型即可用于对多种服装类别进行识别。Specifically, a CNN model is trained by a supervised algorithm based on a plurality of the first training pictures to construct a clothing category recognition model. That is, a CNN model is trained by a deep learning supervised algorithm to learn and recognize clothing categories in the first training clothing picture, construct a clothing category recognition model, and the constructed clothing category recognition model can be used to identify multiple clothing categories.

步骤S104，根据服装类别查找与该服装类别相匹配的服装关键点识别模型。Step S104: Find a clothing keypoint recognition model matching the clothing category according to the clothing category.

在本实施例中，可以由电子设备100的处理器40根据服装类别查找与该服装类别相匹配的服装关键点识别模型。In this embodiment, the processor 40 of the electronic device 100 may find a clothing keypoint recognition model matching the clothing category according to the clothing category.

电子设备100的存储器41中可以预先存储有服装类别与服装关键点识别模型的对应关系表；因此，处理器40通过服装类别查找该对应关系表得到与该服装类别相匹配的服装关键点识别模型。The memory 41 of the electronic device 100 may store a correspondence table between the clothing category and the clothing keypoint recognition model in advance; therefore, the processor 40 searches the correspondence relationship table through the clothing category to obtain a clothing keypoint recognition model that matches the clothing category. .

步骤S105，调用上述服装关键点识别模型识别图片信息中的服装关键点特征信息。In step S105, the clothing keypoint recognition model is called to identify the clothing keypoint feature information in the picture information.

上述的服装关键点信息包括服装关键点类别信息及所述服装关键点位置信息。图3示出了一种服装图片中的关键点特征信息。参照图3示出的T恤及其关键点类别信息表，其中关键点类别信息包括后颈1、左领2、左肩3、左袖外4、左袖内5、左身6、左褶7、右褶8、右身9、右袖内10、右袖外11、右肩12、右领13。The aforementioned clothing keypoint information includes clothing keypoint category information and the clothing keypoint position information. FIG. 3 shows key point feature information in a clothing picture. Referring to FIG. 3, the T-shirt and its key point category information table, where the key point category information includes the back neck 1, left collar 2, left shoulder 3, left sleeve outer 4, left sleeve 5, left body 6, left pleats 7 , Right fold 8, right body 9, right sleeve 10, right sleeve outer 11, right shoulder 12, right collar 13.

预先构建的服装关键点识别模型存储电子设备的存储器中，以便于后续调用及关键点特征信息识别。The pre-built clothing key point recognition model is stored in the memory of the electronic device, so as to facilitate subsequent calls and key point feature information identification.

具体的，服装关键点特征信息包括服装关键点类别信息及服装关键点位置信息，步骤S105包括：Specifically, the clothing keypoint feature information includes clothing keypoint category information and clothing keypoint position information. Step S105 includes:

调用服装关键点识别模型识别图片中的服装关键点类别信息及服装关键点位置信息，服装关键点位置信息可以以像素点的形式表示；或者说关键点位置信息是以像素坐标表示的。The clothing keypoint recognition model is called to identify the clothing keypoint category information and the clothing keypoint position information in the picture. The clothing keypoint position information can be expressed in the form of pixels; or the keypoint position information is expressed in pixel coordinates.

需要说明的是，服装关键点识别模型是通过采用深度学习的监督式算法对CNN模型对训练用服装图片进行训练后构建得到的；上述的训练用服装图片预先标注有关键点特征信息，即训练用服装图片标注有关键点类别信息和关键点位置信息；上述的关键点位置信息 以像素坐标的形式表示，为了提高识别精度，在训练过程中的训练用服装图片的尺寸、角度是一致的；且标注的关键点类别信息和关键点位置信息用于建立关键点数据集，该关键点数据集用于上述服装关键点识别模型的构建过程，具体的，通过监督式算法训练CNN模型识别标注有关键点类别及位置建立的关键点数据集的训练用服装图片的关键类别及其位置。It should be noted that the clothing keypoint recognition model is constructed by training the CNN model on the training clothing pictures by using a supervised algorithm of deep learning; the above training clothing pictures are pre-labeled with keypoint feature information, that is, training The clothing picture is marked with key point category information and key point position information; the above key point position information is expressed in the form of pixel coordinates. In order to improve the recognition accuracy, the size and angle of the training clothing picture during the training process are consistent; And the labeled key point category information and key point position information are used to build a key point data set, and the key point data set is used in the construction process of the clothing key point recognition model described above. Specifically, the CNN model is trained through a supervised algorithm to identify the labeled Keypoint category and location The key category and location of the training clothing picture based on the keypoint dataset.

因此，该方法还包括：构建服装关键点识别模型，该构建服装关键点识别模型包括以下步骤：Therefore, the method further includes: constructing a keypoint recognition model of clothing, and the constructing a keypoint recognition model of clothing includes the following steps:

1、获取第二训练用服装图片；其中第二训练用服装图片为标注有关键点特征信息的图片；第二训练用服装为多件，且，多件所述第二训练用服装对应的关键点特征信息均不同；所述关键点特征信息与所述第二训练用服装的服装类别相对应；1. Obtain a picture of the second training clothing; wherein the picture of the second training clothing is a picture marked with key point feature information; there are multiple pieces of the second training clothing, and the key corresponding to the multiple pieces of the second training clothing The point feature information is different; the key point feature information corresponds to the clothing category of the second training clothing;

具体的，上述的第二训练用服装图片也是在预设拍照环境、由摄像头在预设拍照角度下对预设服装类别的服装拍摄获得的，即摄像头对该预设服装类别的第二训练用服装进行拍摄，得到拍摄图片，然后在该拍摄图片上进行标注，生成第二训练用服装图片。例如可以通过绘图软件在图片上标注，这里不作限定。Specifically, the above-mentioned second training clothing picture is also obtained in a preset photographing environment and the camera photographs the clothing of the preset clothing category at a preset photographing angle, that is, the camera uses the second training clothing for the preset clothing category. The clothing is photographed to obtain a photographed picture, and then annotated on the photographed picture to generate a second training costume picture. For example, you can mark on the picture by drawing software, which is not limited here.

2、基于上述第二训练用服装图片对CNN模型进行训练，以构建得到服装关键点识别模型。2. Train the CNN model based on the second training clothing picture to construct a clothing keypoint recognition model.

具体的，基于该预设服装类别的第二训练用图片通过监督式算法对CNN模型进行训练，构建得到服装关键点识别模型。即通过深度学习的监督式算法训练CNN模型对第二训练用服装图片中的关键点特征信息进行学习和识别，构建服装关键点识别模型。Specifically, the second training picture based on the preset clothing category is used to train the CNN model through a supervised algorithm to construct a clothing keypoint recognition model. That is, a CNN model is trained by a deep learning supervised algorithm to learn and recognize key point feature information in a second training clothing picture, and construct a clothing key point recognition model.

需要说明的是，预设服装类别为多种时，可以训练得到多种服装关键点识别模型，也就是说，服装关键点识别模型与服装类别是一一对应的。It should be noted that when the preset clothing categories are multiple, multiple clothing keypoint recognition models can be trained, that is, the clothing keypoint recognition model and the clothing category have a one-to-one correspondence.

此外，由摄像头拍摄的图片可以同时标注有服装类别和该服装类别的关键点特征信息，以便同时对两种模型进行训练，提高构建效率，即第二训练用图片是在标注有服装类别的训练用图片的基础上进一步标注关键点特征信息的图片，换而言之，服装类别识别模型和服装关键点识别模型采用对预设服装类别拍摄得到的拍摄图片中既标注出服装类别和该服装类别的关键点特征信息(简称为学习图片)；也就是说，训练用图片和第二训练用服装图片可以统一为一张学习图片，该学习图片既可以用来训练得到服装类别识别模型，也可以用于训练得到服装关键点识别模型。In addition, the picture taken by the camera can be labeled with the clothing category and key point feature information of the clothing category at the same time, so that the two models can be trained at the same time to improve the construction efficiency, that is, the second training picture is the training marked with the clothing category Use the picture to further mark the picture of key point feature information. In other words, the clothing category recognition model and the clothing key point recognition model use both the clothing category and the clothing category in the shooting picture obtained by shooting the preset clothing category. Key point feature information (referred to as the learning picture); that is, the training picture and the second training clothing picture can be unified into a learning picture, and the learning picture can be used for training to obtain the clothing category recognition model or Used for training to get clothing keypoint recognition model.

步骤S106，基于服装关键点特征信息进行计算得到待测量服装的第一尺寸信息。Step S106: Perform calculation based on the keypoint feature information of the clothing to obtain the first size information of the clothing to be measured.

在本实施例中，可以由电子设备100的处理器40基于服装关键点特征信息进行计算得到待测量服装的第一尺寸信息。In this embodiment, the processor 40 of the electronic device 100 can calculate the first size information of the clothing to be measured based on the clothing keypoint feature information.

其中所述第一尺寸信息包括所述待测量服装的预设特征部位的像素尺寸；上述的预设 特征部位的像素尺寸包括袖长、胸宽、肩宽、领宽、身长、腰长等。The first size information includes a pixel size of a preset feature part of the clothing to be measured; the pixel size of the preset feature part includes sleeve length, chest width, shoulder width, collar width, body length, waist length, and the like.

具体的，处理器40基于所述服装关键点特征信息以及所述服装类别的预设特征部位的关键点计算规则进行计算得到所述待测量服装的第一尺寸信息。Specifically, the processor 40 calculates the first size information of the clothing to be measured based on the clothing keypoint feature information and the keypoint calculation rule of a preset feature part of the clothing category.

该步骤S106包括以下步骤：This step S106 includes the following steps:

1、处理器40调取与所述服装类别相对应的关键点计算规则；其中所述关键点计算规则包括所述服装类别的预设特征部位的像素尺寸计算规则。1. The processor 40 retrieves a key point calculation rule corresponding to the clothing category; wherein the key point calculation rule includes a pixel size calculation rule of a preset feature part of the clothing category.

2、基于所述服装关键点特征信息以及所述关键点计算规则进行计算得到所述第一尺寸信息。2. Calculate the first size information based on the clothing keypoint feature information and the keypoint calculation rule.

关键点计算规则包括袖长计算规则、胸宽计算规则、肩宽计算规则、领宽计算规则、身长计算规则、腰长计算规则等等，将上述获得服装关键点特征信息结合预先设定的关键点计算规则计算得到待测量服装的预设特征部位的像素尺寸。需要指出的是，关键点计算规则可以称为距离算法，用于计算待测量服装的预设特征部位的距离(像素尺寸)。The key point calculation rules include sleeve length calculation rules, chest width calculation rules, shoulder width calculation rules, collar width calculation rules, body length calculation rules, waist length calculation rules, etc., combining the above-mentioned clothing key point feature information with a predetermined key The point calculation rule is used to calculate the pixel size of the preset feature part of the clothing to be measured. It should be noted that the key point calculation rule can be called a distance algorithm, which is used to calculate the distance (pixel size) of a preset feature part of the clothing to be measured.

下面以图3的T恤为例，腰长计算规则如下：左褶7和右褶8(7、8为关键点类别)的像素坐标(关键点位置)之差，即得到腰长的像素坐标，像素坐标利用勾股定理，计算得到腰长(预设特征部位)的像素尺寸；其他特征部位的计算规则可以参照腰长计算规则，在此不再赘述。The following uses the T-shirt in Figure 3 as an example. The waist length calculation rules are as follows: the difference between the pixel coordinates (keypoint positions) of the left pleats 7 and the right pleats 8 (7, 8 are key point categories), to obtain the pixel coordinates of the waist length The pixel coordinates use the Pythagorean theorem to calculate the pixel size of the waist length (preset feature parts); for the calculation rules of other feature parts, please refer to the waist length calculation rules, which will not be repeated here.

由于不同的服装类别的关键点特征信息以及关键点计算规则是不同的，因此可以将关键点计算规则与服装类别、关键点特征信息对应后存储在存储器中，以方便后续调用进行像素尺寸计算。Because the key point feature information and key point calculation rules are different for different clothing categories, the key point calculation rules can be stored in the memory after corresponding to the clothing category and key point feature information, so as to facilitate subsequent call for pixel size calculation.

步骤S107，基于第一尺寸信息和比率系数计算得到第二尺寸信息。Step S107, calculating the second size information based on the first size information and the ratio coefficient.

在本实施例中，可以由电子设备100的处理器40基于第一尺寸信息和比率系数计算得到第二尺寸信息。In this embodiment, the second size information may be calculated by the processor 40 of the electronic device 100 based on the first size information and the ratio coefficient.

其中，第二尺寸信息包括待测量服装的所述预设特征部位的实际尺寸。Wherein, the second size information includes an actual size of the preset feature part of the clothing to be measured.

具体的，处理器40将第一尺寸信息和比率系数相乘即可计算得到第二尺寸信息。Specifically, the processor 40 can calculate the second size information by multiplying the first size information and the ratio coefficient.

通过该步骤S107能够实现将图片信息中的像素尺寸经过尺寸转换得到真实尺寸，完成服装的尺寸测量。Through this step S107, the pixel size in the picture information can be converted to obtain the true size, and the size measurement of the clothing can be completed.

本申请实施例提供的基于图片的服装尺寸测量方法包括：接收用户上传的图片信息，图片信息为包括预设参照物和待测量服装的图片；图片信息是在预设拍照环境、由摄像头在预设拍照角度下获得的；获取图片信息中的预设参照物的像素尺寸数据以及实际尺寸数据，基于像素尺寸数据和实际尺寸数据生成比率系数；识别图片信息中的待测量服装的服装类别；根据服装类别查找与服装类别相匹配的服装关键点识别模型；调用服装关键点识别模型识别图片信息中的服装关键点特征信息；基于服装关键点特征信息进行计算得到待 测量服装的第一尺寸信息；其中第一尺寸信息包括待测量服装的预设特征部位的像素尺寸；基于第一尺寸信息和比率系数计算得到第二尺寸信息；其中第二尺寸信息包括待测量服装的预设特征部位的实际尺寸。因此，本申请实施例提供的技术方案，缓解了现有技术中依靠人工的方式存在的效率低下的问题，能够提高测量效率。该方法基于用户输入的图片，对图片进行测量得到服装的多个部位的尺寸测量值(真实值)，相比于现有技术中依靠人工的方式存在的效率低下的问题，能够提高测量效率；该方法在预测服装关键点结合了AR编码识别以及AI视觉技术，识别准确度高，测试简单、方便、快速，大大提高了服装尺寸的测量效率，节省了人力、时间成本。此外，该方法还具有以下优点：1)可适配多种服装类别，不需要特制模型；2)可识别个别服装部位并测量相关尺寸；3)高效率，可以快速拍照多件衣服；4)简单部署，除了摄像头及基本道具不需要特定设备，5)误差小，能够达到商业要求的精准度。The picture-based clothing size measurement method provided in the embodiment of the present application includes: receiving picture information uploaded by a user, the picture information is a picture including a preset reference object and a garment to be measured; the picture information is in a preset photographing environment, Obtained at a photographing angle; obtain pixel size data and actual size data of a preset reference object in the picture information, and generate a ratio coefficient based on the pixel size data and the actual size data; identify the clothing category of the clothing to be measured in the picture information; Find the clothing keypoint recognition model that matches the clothing category; call the clothing keypoint recognition model to identify the clothing keypoint feature information in the picture information; perform calculations based on the clothing keypoint feature information to obtain the first size information of the clothing to be measured; The first size information includes the pixel size of the preset feature part of the clothing to be measured; the second size information is calculated based on the first size information and the ratio coefficient; wherein the second size information includes the actual size of the preset feature part of the clothing to be measured . Therefore, the technical solutions provided in the embodiments of the present application alleviate the problem of low efficiency in the prior art relying on manual methods, and can improve measurement efficiency. This method is based on a picture input by a user, and measures the pictures to obtain the dimensional measurement values (real values) of multiple parts of the garment. Compared with the problem of low efficiency of relying on manual methods in the prior art, the measurement efficiency can be improved; This method combines AR code recognition and AI vision technology at key points for predicting clothing. The recognition accuracy is high, the test is simple, convenient and fast, which greatly improves the measurement efficiency of clothing size and saves manpower and time costs. In addition, this method also has the following advantages: 1) can be adapted to a variety of clothing categories, no special model is required; 2) can identify individual clothing parts and measure related sizes; 3) high efficiency, can quickly take pictures of multiple pieces of clothing; 4) Simple deployment, in addition to the camera and basic props do not require specific equipment, 5) small error, can achieve the accuracy required by business.

需要说明的是，步骤S102仅为描述方便使用，不代表其在该方法中的先后顺序；例如该步骤S102可以在步骤S101和步骤S107之间的任意步骤之前或之后；当然该步骤S102也可以与步骤S103同时进行；因此，上述步骤编号不应理解为对本申请造成的限制。It should be noted that step S102 is only for convenience of description, and does not represent its sequence in the method; for example, step S102 can be before or after any step between step S101 and step S107; of course, step S102 can also be Simultaneously with step S103; therefore, the above step numbers should not be construed as a limitation on the present application.

进一步的，该方法还包括输出步骤S108。Further, the method further includes an output step S108.

步骤S108，将第二尺寸信息输出。In step S108, the second size information is output.

在本实施例中，可以由电子设备100的通信接口43输出第二尺寸信息。In this embodiment, the second size information may be output by the communication interface 43 of the electronic device 100.

具体的，将所述第二尺寸信息输出至ERP***或者BOM数据库，由ERP***或者BOM数据库进行存储。Specifically, the second size information is output to an ERP system or a BOM database, and is stored by the ERP system or the BOM database.

另外，上述待测量服装图片、服装类别、关键点特征信息以及第二尺寸信息均存储在ERP***或者BOM数据库中。In addition, the clothing picture to be measured, the clothing category, key point feature information, and the second size information are all stored in the ERP system or the BOM database.

这里需要进行说明的是，ERP***是企业资源计划(Enterprise Resource Planning)的简称，在信息技术基础上，集信息技术与先进管理思想于一身，其核心思想是供应链管理。ERP***从供应链范围去优化企业的资源，优化了现代企业的运行模式，数据在各业务***之间高度共享，所有源数据只需在某一个***中输入一次，保证了数据的一致性。What needs to be explained here is that ERP system is the abbreviation of Enterprise Resource Planning. Based on information technology, it integrates information technology and advanced management thinking. Its core idea is supply chain management. The ERP system optimizes the company's resources from the scope of the supply chain and optimizes the operating model of modern enterprises. Data is highly shared between business systems. All source data need only be entered once in a certain system to ensure data consistency.

BOM数据库是以数据格式来描述产品结构的文件，是计算机可以识别的产品结构数据文件，也是ERP的主导文件。BOM数据库使***识别产品结构，也是联系与沟通企业各项业务的纽带。BOM数据库是计算机识别物料的基础依据，是编制计划的依据，是配套和领料的依据，是采购和外协的依据，能够使设计系列化、标准化、通用化。The BOM database is a file that describes the product structure in a data format. It is a product structure data file that can be identified by a computer, and it is also the leading file for ERP. The BOM database enables the system to identify the product structure, and is also the link between the various businesses of the enterprise. The BOM database is the basis for computer identification of materials, the basis for the preparation of plans, the basis for matching and picking, the basis for procurement and outsourcing, and it can serialize, standardize and generalize designs.

为了便于理解，下面结合图4对本申请实施例提供的基于图片的服装尺寸测量方法的应用场景进行简要说明：In order to facilitate understanding, an application scenario of the picture-based clothing size measurement method provided in the embodiment of the present application is briefly described below with reference to FIG. 4:

1、图片输入：电子设备(如计算机终端、服务器等)通过API接口接收输入的图片； 图片包括待测量的服装和打印有AR代码的矩形卡纸；1. Picture input: electronic devices (such as computer terminals, servers, etc.) receive input pictures through the API interface; pictures include clothing to be measured and rectangular cardboard with AR code printed;

2、服装分类：电子设备对输入的图片进行服装类别识别，完成服装分类；2. Clothing classification: The electronic device recognizes the clothing category of the input picture and completes the clothing classification;

3、服装关键点检测：电子设备对分类后的服装进行关键点检测，识别出关键点；3. Clothing key point detection: electronic equipment performs key point detection on classified clothing to identify key points;

4、距离算法：电子设备利用距离算法对关键点进行距离计算，得到服装的相应特征部位的像素尺寸；4. Distance algorithm: The electronic device uses the distance algorithm to calculate the distance of the key points to obtain the pixel size of the corresponding characteristic parts of the clothing;

5、AR代码识别：电子设备在进行服装分类的同时，利用AR编码识别***识别已设置的固定编码；5. AR code recognition: Electronic equipment uses the AR code recognition system to identify fixed codes that have been set while classifying clothing;

6、像素尺寸算法：用图像识别算法计算参照物的边框像素距离，再参考已设置好的实际卡纸尺寸，利用像素尺寸算法计算出毫米对像素比率；6. Pixel size algorithm: Use the image recognition algorithm to calculate the pixel distance of the frame of the reference object, and then refer to the actual paper jam size that has been set, and use the pixel size algorithm to calculate the millimeter to pixel ratio;

7、尺寸转换：基于距离算法得到的服装的相应特征部位的像素尺寸以及像素尺寸算法计算出的毫米对像素比率，将像素尺寸转换为实际尺寸；7. Size conversion: Based on the pixel size of the corresponding feature parts of the clothing obtained by the distance algorithm and the millimeter-to-pixel ratio calculated by the pixel size algorithm, the pixel size is converted to the actual size;

8、尺寸输出：将经尺寸转换得到的服装的实际尺寸输出。8. Size output: Output the actual size of the garment obtained through size conversion.

实施例二：Embodiment two:

如图5所示，本申请实施例提供了一种基于图片的服装尺寸测量装置，该服装尺寸测量装置可以为电子设备100的存储器41中的机器可执行的计算机程序，该服装尺寸测量装置包括可以被处理器40执行的一个或多个功能模块。As shown in FIG. 5, an embodiment of the present application provides a picture-based clothing size measurement device. The clothing size measurement device may be a machine-executable computer program in the memory 41 of the electronic device 100. The clothing size measurement device includes One or more functional modules that can be executed by the processor 40.

从功能上划分，服装尺寸测量装置可以包括：图片接收模块100、比例获取模块200、类别识别模块300、查找模块400、关键点识别模块500、第一计算模块600和第二计算模块700。Functionally, the clothing size measurement device may include: a picture receiving module 100, a scale acquisition module 200, a category identification module 300, a search module 400, a key point identification module 500, a first calculation module 600, and a second calculation module 700.

其中，图片接收模块100配置成接收用户上传的图片信息，所述图片信息为包括预设参照物和待测量服装的图片；所述图片信息是在预设拍照环境、由摄像头在预设拍照角度下获得的；The picture receiving module 100 is configured to receive picture information uploaded by a user, where the picture information is a picture including a preset reference object and clothing to be measured; the picture information is in a preset photographing environment, and the camera is at a preset photographing angle Obtained under

具体的，所述预设参照物为预设形状的卡纸；所述卡纸上包括有服装识别码，所述服装识别码与所述电子设备所述服装识别码仅能够被所述电子设备识别；所述电子设备仅能够识别所述服装识别码；Specifically, the preset reference object is a paperboard of a preset shape; the paperboard includes a clothing identification code, and the clothing identification code and the electronic device can be used only by the electronic device. Identification; the electronic device can only identify the clothing identification code;

所述预设拍照环境是通过摄影棚控制调节得到；The preset photographing environment is obtained through studio control adjustment;

所述预设拍照角度是通过摄影架控制调节得到，以减少摄像头水平拍照时角度畸变；所述摄影架构造成设置所述摄像头。The preset photographing angle is obtained through control adjustment of a photographic frame, so as to reduce the angle distortion when the camera is horizontally photographed; the photographing framework causes the camera to be set.

比例获取模块200配置成获取所述图片信息中的所述预设参照物的像素尺寸数据以及实际尺寸数据，基于所述像素尺寸数据和所述实际尺寸数据生成比率系数；The scale obtaining module 200 is configured to obtain pixel size data and actual size data of the preset reference object in the picture information, and generate a ratio coefficient based on the pixel size data and the actual size data;

类别识别模块300配置成识别所述图片信息中的所述待测量服装的服装类别；A category identification module 300 configured to identify a clothing category of the clothing to be measured in the picture information;

查找模块400配置成根据所述服装类别查找与所述服装类别相匹配的服装关键点识别模型；The search module 400 is configured to find a clothing keypoint recognition model matching the clothing category according to the clothing category;

关键点识别模块500配置成调用所述服装关键点识别模型识别所述图片信息中的服装关键点特征信息；The key point recognition module 500 is configured to call the clothing key point recognition model to identify clothing key point feature information in the picture information;

第一计算模块600配置成基于所述服装关键点特征信息进行计算得到所述待测量服装的第一尺寸信息；其中所述第一尺寸信息包括所述待测量服装的预设特征部位的像素尺寸；The first calculation module 600 is configured to perform calculation based on the keypoint feature information of the clothing to obtain first size information of the clothing to be measured; wherein the first size information includes a pixel size of a preset feature part of the clothing to be measured ;

第二计算模块700配置成基于所述第一尺寸信息和所述比率系数计算得到第二尺寸信息；其中所述第二尺寸信息包括所述待测量服装的所述预设特征部位的实际尺寸。The second calculation module 700 is configured to calculate and obtain second size information based on the first size information and the ratio coefficient; wherein the second size information includes an actual size of the preset feature part of the clothing to be measured.

进一步的，该装置还包括输出模块800，配置成将所述第二尺寸信息输出。Further, the device further includes an output module 800 configured to output the second size information.

进一步的，关键点识别模块500具体配置成调用所述服装关键点识别模型识别所述图片中的服装关键点类别信息及所述服装关键点位置信息。Further, the keypoint recognition module 500 is specifically configured to call the clothing keypoint recognition model to identify the clothing keypoint category information and the clothing keypoint position information in the picture.

进一步的，第一计算模块600具体配置成调取与所述服装类别相对应的关键点计算规则；其中所述关键点计算规则包括所述服装类别的预设特征部位的像素尺寸计算规则；基于所述服装关键点特征信息以及所述关键点计算规则进行计算得到所述第一尺寸信息。Further, the first calculation module 600 is specifically configured to call a key point calculation rule corresponding to the clothing category; wherein the key point calculation rule includes a pixel size calculation rule of a preset feature part of the clothing category; based on The clothing keypoint feature information and the keypoint calculation rule are calculated to obtain the first size information.

本申请实施例提供的基于图片的服装尺寸测量装置，与上述实施例提供的基于图片的服装尺寸测量方法具有相同的技术特征，所以也能解决相同的技术问题，达到相同的技术效果。The image-based clothing size measurement device provided in the embodiment of the present application has the same technical features as the image-based clothing size measurement method provided in the foregoing embodiment, so it can also solve the same technical problems and achieve the same technical effect.

需要说明的是，本说明书中的各个实施例均采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似的部分互相参见即可。本申请实施例所提供的基于图片的服装尺寸测量方法及装置，其实现原理及产生的技术效果和前述方法实施例相同，为简要描述，装置实施例部分未提及之处，可参考前述方法实施例中相应内容。It should be noted that each embodiment in this specification is described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments refer to each other. can. The image-based clothing size measurement method and device provided in the embodiments of the present application have the same implementation principles and technical effects as the previous method embodiments. For a brief description, the parts not mentioned in the device embodiments can be referred to the foregoing methods Corresponding content in the examples.

在本申请所提供的几个实施例中，应该理解到，所揭露的装置和方法，也可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，附图中的流程图和框图显示了根据本申请的多个实施例的装置、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或代码的一部分，所述模块、程序段或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现方式中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专 用的基于硬件的***来实现，或者可以用专用硬件与计算机指令的组合来实现。In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may also be implemented in other ways. The device embodiments described above are only schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functions, and functions of devices, methods, and computer program products according to various embodiments of the present application. operating. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, which contains one or more components for implementing a specified logical function Executable instructions. It should also be noted that in some alternative implementations, the functions marked in the blocks may also occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented in a dedicated hardware-based system that performs the specified function or action. , Or it can be implemented with a combination of dedicated hardware and computer instructions.

另外，在本申请各个实施例中的各功能模块或单元可以集成在一起形成一个独立的部分，也可以是各个模块单独存在，也可以两个或两个以上模块集成形成一个独立的部分。In addition, the functional modules or units in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

所述功能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，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application is essentially a part that contributes to the existing technology or a part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present application. The foregoing storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other media that can store program codes .

需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序，也不能理解为指示或暗示相对重要性。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。It should be noted that in this article, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that these entities or operations The existence of any such actual relationship or order cannot be understood as indicating or implying relative importance. Moreover, the terms "including", "comprising", or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or device that includes a series of elements includes not only those elements but also those that are not explicitly listed Or other elements inherent to such a process, method, article, or device. Without more restrictions, the elements defined by the sentence "including a ..." do not exclude the existence of other identical elements in the process, method, article, or equipment including the elements.

以上所述仅为本申请的优选实施例而已，并不用于限制本申请，对于本领域的技术人员来说，本申请可以有各种更改和变化。凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步定义和解释。The above description is only a preferred embodiment of the present application, and is not intended to limit the present application. For those skilled in the art, this application may have various modifications and changes. Any modification, equivalent replacement, or improvement made within the spirit and principle of this application shall be included in the protection scope of this application. It should be noted that similar reference numerals and letters indicate similar items in the following drawings, so once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings.

最后应说明的是：以上所述实施例，仅为本申请的具体实施方式，用以说明本申请的技术方案，而非对其限制，本申请的保护范围并不局限于此，尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，其依然可以对前述实施例所记载的技术方案进行修改或可轻易想到变化，或者对其中部分技术特征进行等同替换；而这些修改、变化或者替换，并不使相应技术方案的本质脱离本申请实施例技术方案的精神和范围，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应所述以权利要求的保护范围为准。Finally, it should be noted that the above-mentioned embodiments are only specific implementations of the present application, and are used to describe the technical solutions of the present application, but not to limit them. The scope of protection of the present application is not limited to this. The embodiments describe this application in detail. Those of ordinary skill in the art should understand that anyone skilled in the art can still modify the technical solutions described in the foregoing embodiments within the technical scope disclosed in this application. Or you can easily think of changes, or equivalent replacements of some of the technical features; and these modifications, changes, or replacements do not deviate the essence of the corresponding technical solution from the spirit and scope of the technical solutions of the embodiments of the present application, which should be covered in this application Within the scope of protection. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

工业实用性Industrial applicability

本申请实施例提供的基于图片的服装尺寸测量方法、装置、电子设备及计算机可读存储介质，其中，该基于图片的服装尺寸测量方法包括：接收用户上传的图片信息，图片信息为包括预设参照物和待测量服装的图片；图片信息是在预设拍照环境、由摄像头在预设拍照角度下获得的；获取图片信息中的预设参照物的像素尺寸数据以及实际尺寸数据，基于像素尺寸数据和实际尺寸数据生成比率系数；识别图片信息中的待测量服装的服装类别；根据服装类别查找与服装类别相匹配的服装关键点识别模型；调用服装关键点识别模型识别图片信息中的服装关键点特征信息；基于服装关键点特征信息进行计算得到待测量服装的第一尺寸信息；其中第一尺寸信息包括待测量服装的预设特征部位的像素尺寸；基于第一尺寸信息和比率系数计算得到第二尺寸信息；其中第二尺寸信息包括待测量服装的预设特征部位的实际尺寸。因此，本申请实施例提供的技术方案，缓解了现有技术中依靠人工的方式存在的效率低下的问题，能够提高测量效率。该方法基于用户输入的图片，对图片进行测量得到服装的多个部位的尺寸测量值，相比于现有技术中依靠人工的方式存在的效率低下的问题，能够提高测量效率，该方法简单、方便、快速，大大提高了服装尺寸的测量效率，节省了人力、时间成本。The image-based clothing size measurement method, device, electronic device, and computer-readable storage medium provided in the embodiments of the present application, wherein the image-based clothing size measurement method includes: receiving picture information uploaded by a user, and the picture information includes a preset Pictures of the reference object and the clothing to be measured; the picture information is obtained in the preset photographing environment and by the camera at the preset photographing angle; the pixel size data and actual size data of the preset reference object in the picture information are obtained, based on the pixel size Data and actual size data to generate a ratio coefficient; identify the clothing category of the clothing to be measured in the picture information; find the clothing keypoint recognition model that matches the clothing category according to the clothing category; call the clothing keypoint recognition model to identify the clothing key in the picture information Point feature information; calculation based on the key point feature information of the clothing to obtain the first size information of the clothing to be measured; wherein the first size information includes the pixel size of the predetermined feature part of the clothing to be measured; calculated based on the first size information and the ratio coefficient Second size information; The second feature dimension information includes a preset section of the actual dimensions of the garment to be measured. Therefore, the technical solutions provided in the embodiments of the present application alleviate the problem of low efficiency in the prior art relying on manual methods, and can improve measurement efficiency. This method is based on a picture input by a user, and measures the pictures to obtain the size measurement values of multiple parts of the garment. Compared with the problem of inefficiency of relying on manual methods in the prior art, the method can improve measurement efficiency. Convenient and fast, greatly improving the measurement efficiency of clothing size, saving labor and time costs.

Claims (13)

1. 一种基于图片的服装尺寸测量方法，其特征在于，包括：A picture-based clothing size measurement method, comprising:

   接收用户上传的图片信息，所述图片信息为包括预设参照物和待测量服装的图片；所述图片信息是在预设拍照环境、由摄像头在预设拍照角度下获得的；Receiving picture information uploaded by a user, the picture information being a picture including a preset reference object and clothing to be measured; the picture information is obtained in a preset photographing environment and a camera at a preset photographing angle;

   获取所述图片信息中的所述预设参照物的像素尺寸数据以及实际尺寸数据，基于所述像素尺寸数据和所述实际尺寸数据生成比率系数；Acquiring pixel size data and actual size data of the preset reference object in the picture information, and generating a ratio coefficient based on the pixel size data and the actual size data;

   识别所述图片信息中的所述待测量服装的服装类别；Identifying a clothing category of the clothing to be measured in the picture information;

   根据所述服装类别查找与所述服装类别相匹配的服装关键点识别模型；Find a clothing keypoint recognition model matching the clothing category according to the clothing category;

   调用所述服装关键点识别模型识别所述图片信息中的服装关键点特征信息；Calling the clothing keypoint recognition model to identify clothing keypoint feature information in the picture information;

   基于所述服装关键点特征信息进行计算得到所述待测量服装的第一尺寸信息；其中所述第一尺寸信息包括所述待测量服装的预设特征部位的像素尺寸；Calculate the first size information of the clothing to be measured based on the clothing keypoint feature information; wherein the first size information includes a pixel size of a preset feature part of the clothing to be measured;

   基于所述第一尺寸信息和所述比率系数计算得到第二尺寸信息；其中所述第二尺寸信息包括所述待测量服装的所述预设特征部位的实际尺寸。The second size information is calculated based on the first size information and the ratio coefficient; wherein the second size information includes an actual size of the preset feature part of the clothing to be measured.
2. 根据权利要求1所述的基于图片的服装尺寸测量方法，其特征在于，所述图片信息中包括平铺状态下的待测量服装的图片。The method for measuring the size of a garment based on a picture according to claim 1, wherein the picture information includes a picture of the garment to be measured in a tiled state.
3. 根据权利要求1或2所述的基于图片的服装尺寸测量方法，其特征在于，所述调用所述服装关键点识别模型识别所述图片中的服装关键点特征信息，包括：The method for measuring a size of a garment based on a picture according to claim 1 or 2, wherein the invoking the garment keypoint recognition model to identify the keypoint feature information of the garment in the picture comprises:

   调用所述服装关键点识别模型识别所述图片中的服装关键点类别信息及所述服装关键点位置信息。Calling the clothing keypoint recognition model to identify clothing keypoint category information and the clothing keypoint position information in the picture.
4. 根据权利要求1-3中任一项所述的基于图片的服装尺寸测量方法，其特征在于，所述预设参照物为预设形状的卡纸；所述卡纸上包括有服装识别码，所述服装识别码仅能够被所述电子设备识别；所述电子设备仅能够识别所述服装识别码。The method for measuring a garment size based on a picture according to any one of claims 1 to 3, wherein the preset reference object is a cardboard with a preset shape; the cardboard includes a clothing identification code, The clothing identification code can only be recognized by the electronic device; the electronic device can only identify the clothing identification code.
5. 根据权利要求1-4中任一项所述的基于图片的服装尺寸测量方法，其特征在于，所述预设拍照环境是通过摄影棚控制调节得到。The method for measuring a size of a garment based on a picture according to any one of claims 1-4, wherein the preset photographing environment is obtained through a studio control adjustment.
6. 根据权利要求1至5中任一项所述的基于图片的服装尺寸测量方法，其特征在于，所述预设拍照角度是通过摄影架控制调节得到，以减少摄像头水平拍照时角度畸变；所述摄影架构造成设置所述摄像头。The method for measuring the size of a garment based on a picture according to any one of claims 1 to 5, wherein the preset photographing angle is obtained through control of a photographic frame to reduce angle distortion when the camera is horizontally photographing; The photographic architecture results in setting up the camera.
7. 根据权利要求1至6中任一项所述的基于图片的服装尺寸测量方法，其特征在于，所述方法还包括：The picture-based clothing size measurement method according to any one of claims 1 to 6, wherein the method further comprises:

   将所述第二尺寸信息输出。And outputting the second size information.
8. 根据权利要求1至7中任一项所述的基于图片的服装尺寸测量方法，其特征在于，所述基于所述服装关键点特征信息进行计算得到所述待测量服装的第一尺寸信息，包 括：The method for measuring the size of a garment based on a picture according to any one of claims 1 to 7, wherein the first dimension information of the garment to be measured is obtained by calculating based on the key feature information of the garment, comprising: :

   调取与所述服装类别相对应的关键点计算规则；其中所述关键点计算规则包括所述服装类别的预设特征部位的像素尺寸计算规则；Calling a key point calculation rule corresponding to the clothing category; wherein the key point calculation rule includes a pixel size calculation rule of a preset feature part of the clothing category;

   基于所述服装关键点特征信息以及所述关键点计算规则进行计算得到所述第一尺寸信息。Calculate and obtain the first size information based on the clothing keypoint feature information and the keypoint calculation rule.
9. 根据权利要求1-8任意一项所述的基于图片的服装尺寸测量方法，其特征在于，所述方法还包括：The picture-based clothing size measurement method according to any one of claims 1-8, wherein the method further comprises:

   获取多个第一训练用服装图片，其中，所述第一训练用服装图片为标注有服装类别的图片；Obtaining a plurality of first training clothing pictures, wherein the first training clothing pictures are pictures marked with a clothing category;

   基于所述第一训练用服装图片对卷积神经网络模型进行机器学习训练，以构建得到所述服装类别识别模型。Machine learning training is performed on the convolutional neural network model based on the first training clothing picture to construct and obtain the clothing category recognition model.
10. 根据权利要求1-8任意一项所述的基于图片的服装尺寸测量方法，其特征在于，所述方法还包括：The picture-based clothing size measurement method according to any one of claims 1-8, wherein the method further comprises:

    获取多个第二训练用服装图片，所述第二训练用服装图片为标注有服装关键点特征信息的图片；Obtaining a plurality of second training clothing pictures, where the second training clothing pictures are pictures labeled with key point feature information of the clothing;

    基于所述第二训练用服装图片对卷积神经网络模型进行机器学习训练，以构建得到所述服装关键点识别模型。Machine learning training is performed on the convolutional neural network model based on the second training clothing picture to construct the clothing keypoint recognition model.
11. 一种基于图片的服装尺寸测量装置，其特征在于，包括：A picture-based clothing size measuring device, comprising:

    图片接收模块，配置成接收用户上传的图片信息，所述图片信息为包括预设参照物和待测量服装的图片；所述图片信息是在预设拍照环境、由摄像头在预设拍照角度下获得的；A picture receiving module configured to receive picture information uploaded by a user, the picture information being a picture including a preset reference object and clothing to be measured; the picture information is obtained in a preset photographing environment and by a camera at a preset photographing angle of;

    比例获取模块，配置成获取所述图片信息中的所述预设参照物的像素尺寸数据以及实际尺寸数据，基于所述像素尺寸数据和所述实际尺寸数据生成比率系数；A scale obtaining module configured to obtain pixel size data and actual size data of the preset reference object in the picture information, and generate a ratio coefficient based on the pixel size data and the actual size data;

    类别识别模块，配置成识别所述图片信息中的所述待测量服装的服装类别；A category identification module configured to identify a clothing category of the clothing to be measured in the picture information;

    查找模块，配置成根据所述服装类别查找与所述服装类别相匹配的服装关键点识别模型；A search module configured to find a clothing keypoint recognition model matching the clothing category according to the clothing category;

    关键点识别模块，配置成调用所述服装关键点识别模型识别所述图片信息中的服装关键点特征信息；A key point recognition module configured to call the clothing key point recognition model to identify clothing key point feature information in the picture information;

    第一计算模块，配置成基于所述服装关键点特征信息进行计算得到所述待测量服装的第一尺寸信息；其中所述第一尺寸信息包括所述待测量服装的预设特征部位的像素尺寸；A first calculation module configured to perform calculation based on the clothing keypoint feature information to obtain first size information of the clothing to be measured; wherein the first size information includes a pixel size of a preset feature portion of the clothing to be measured ;

    第二计算模块，配置成基于所述第一尺寸信息和所述比率系数计算得到第二尺寸 信息；其中所述第二尺寸信息包括所述待测量服装的所述预设特征部位的实际尺寸。A second calculation module is configured to calculate and obtain second size information based on the first size information and the ratio coefficient; wherein the second size information includes an actual size of the preset feature part of the clothing to be measured.
12. 一种电子设备，其特征在于，包括存储器以及处理器，所述存储器配置成存储支持处理器执行权利要求1至10任一项所述方法的程序，所述处理器配置成执行所述存储器中存储的程序。An electronic device, comprising a memory and a processor, wherein the memory is configured to store a program that supports the processor to execute the method according to any one of claims 1 to 10, and the processor is configured to execute the memory. Stored programs.
13. 一种计算机可读存储介质，计算机可读存储介质上存储有计算机程序，其特征在于，计算机程序被处理器运行时执行上述权利要求1至10任一项所述方法的步骤。A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is run by a processor, the steps of the method according to any one of claims 1 to 10 are performed.

Images