WO2019096179A1 - 服装logo效果检测方法、装置及电子设备 - Google Patents

服装logo效果检测方法、装置及电子设备 Download PDF

Info

Publication number
WO2019096179A1
WO2019096179A1 PCT/CN2018/115496 CN2018115496W WO2019096179A1 WO 2019096179 A1 WO2019096179 A1 WO 2019096179A1 CN 2018115496 W CN2018115496 W CN 2018115496W WO 2019096179 A1 WO2019096179 A1 WO 2019096179A1
Authority
WO
WIPO (PCT)
Prior art keywords
logo
learning model
clothing
image
detected
Prior art date
Application number
PCT/CN2018/115496
Other languages
English (en)
French (fr)
Inventor
斯科特·马修·罗伯特
黄鼎隆
傅恺
Original Assignee
深圳码隆科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳码隆科技有限公司 filed Critical 深圳码隆科技有限公司
Publication of WO2019096179A1 publication Critical patent/WO2019096179A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G06F18/2148—Generating training patterns; Bootstrap methods, e.g. bagging or boosting characterised by the process organisation or structure, e.g. boosting cascade
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30124—Fabrics; Textile; Paper
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30204—Marker
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/09—Recognition of logos

Definitions

  • the present application relates to the field of image recognition technologies, and in particular, to a method, device, and electronic device for detecting a clothing LOGO effect.
  • the purpose of the present application is to provide a method, a device, and an electronic device for detecting a clothing LOGO effect, which can identify a clothing LOGO image through a deep learning model based on an anti-neural network, and generate an effect diagram under different processes.
  • a clothing LOGO effect which can identify a clothing LOGO image through a deep learning model based on an anti-neural network, and generate an effect diagram under different processes.
  • an embodiment of the present application provides a method for detecting a clothing LOGO effect, including:
  • the LOGO feature is identified by a classifier based on the preset depth learning model training, and the recognition result of the LOGO image to be detected is generated; the recognition result includes: an effect diagram corresponding to the plurality of processes.
  • the embodiment of the present application provides a third possible implementation manner of the first aspect, wherein the classifier based on the preset deep learning model training is obtained by:
  • the embodiment of the present application provides a fourth possible implementation manner of the first aspect, wherein the LOGO feature is identified by using a classifier based on a preset deep learning model to generate an image of the LOGO image to be detected.
  • the results include:
  • the data format includes a binary data format
  • the LOGO feature is converted into a binary data format, and then a classifier based on the preset depth learning model training is input to generate a recognition result corresponding to the LOGO image to be detected.
  • the embodiment of the present application provides a fifth possible implementation manner of the first aspect, wherein the LOGO feature is identified by using a classifier based on a preset depth learning model, and the LOGO image of the garment to be detected is generated. After identifying the results, it also includes:
  • the recognition result of the clothing LOGO image to be inspected is saved as a new clothing LOGO sample data in the clothing LOGO sample database.
  • the embodiment of the present application provides a clothing LOGO effect detecting apparatus, where the apparatus includes:
  • An image acquisition module configured to acquire a LOGO image of the garment to be inspected
  • the feature extraction module is configured to extract a LOGO feature of the LOGO image to be detected according to the preset depth learning model, and the preset depth learning model includes a deep learning model based on the anti-neural network;
  • the garment LOGO identification module is configured to identify the LOGO feature by using a classifier based on the preset depth learning model training, and generate a recognition result of the LOGO image to be detected; the recognition result includes: an effect diagram corresponding to the plurality of processes.
  • an embodiment of the present application provides an electronic device, including a memory and a processor, where the computer stores a computer program executable on a processor, and when the processor executes the computer program, the steps of the method described in the first aspect are implemented. .
  • the embodiment of the present application further provides a computer readable medium having a processor-executable non-volatile program code, the program code causing a processor to perform the method as described in the first aspect.
  • the LOGO image of the garment to be detected is first obtained; and then the LOGO feature of the LOGO image of the garment to be detected is extracted according to the preset depth learning model, wherein the preset depth learning model includes an anti-neural nerve.
  • the deep learning model of the network finally, the LOGO feature is identified by the classifier based on the preset deep learning model training, and the recognition result of the LOGO image of the garment to be detected is generated; the recognition result includes: an effect diagram corresponding to various processes.
  • the method can identify the clothing LOGO image through the deep learning model based on the anti-neural network, and generate the effect diagram under different processes for the reference of the staff, without the need to proof the clothing LOGO, saving a lot of time and cost.
  • FIG. 2 is a flowchart of another method for detecting a clothing LOGO effect according to Embodiment 1 of the present application;
  • FIG. 3 is a flowchart of another method for detecting a clothing LOGO effect according to Embodiment 1 of the present application;
  • FIG. 4 is a flowchart of another method for detecting a LOGO effect of a garment according to Embodiment 1 of the present application;
  • FIG. 5 is a schematic structural diagram of a clothing LOGO effect detecting apparatus according to Embodiment 2 of the present application.
  • FIG. 6 is a schematic structural diagram of an electronic device according to Embodiment 3 of the present application.
  • the existing clothing LOGO effect detection method needs to produce samples of various processes for the same LOGO, which is not only time-consuming and laborious, but also inefficient, especially for complex processes, which takes time and cost. Very big. Based on this, the clothing LOGO effect detection method, device and electronic device provided by the embodiments of the present application can identify the clothing LOGO image through the deep learning model based on the anti-neural network, and generate the effect diagrams under different processes for the staff. For reference, there is no need to proof the clothing LOGO, saving a lot of time and cost.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • the embodiment of the present application provides a garment LOGO effect detecting method, which can be applied to the field of garment production.
  • the method includes:
  • the designed clothing LOGO is photographed by the camera, thereby collecting the LOGO image of the garment to be inspected.
  • S102 Extract a LOGO feature of the to-be-detected clothing LOGO image according to the preset depth learning model, where the preset depth learning model includes a deep learning model based on the anti-neural network.
  • the deep learning model based on the anti-neural network is trained by the clothing LOGO sample data exceeding a certain threshold, and the clothing LOGO sample data includes the pictures corresponding to the clothing LOGO produced by different processes.
  • the above various deep learning models based on the anti-neural network can be implemented by the Caffe deep learning framework.
  • the more the number of clothing LOGO images in the clothing LOGO sample data the better, the more types of crafts, such as thick board glue, digital printing, appliqué embroidery, sequin embroidery, three-dimensional glue process, jacquard, three-dimensional thick Glue, hollow, patch seams and/or printed patches.
  • crafts such as thick board glue, digital printing, appliqué embroidery, sequin embroidery, three-dimensional glue process, jacquard, three-dimensional thick Glue, hollow, patch seams and/or printed patches.
  • the more data the better the versatility of the training-derived deep learning model based on the neural network. This is conducive to the accurate identification of the clothing LOGO image to be detected later, and presents the renderings under various processes, providing reference for the LOGO effect for the relevant staff.
  • the step S102 specifically includes: performing the feature training in the plurality of base layers included in the preset depth learning model as the input image as the input image, and extracting the full connection layer or other designations in the multiple integrations after the training is completed.
  • the feature vector outputted by the base layer is used as the corresponding LOGO feature in the LOGO image of the garment to be inspected.
  • S103 Identifying the LOGO feature by using a classifier trained according to the preset depth learning model, and generating a recognition result of the LOGO image to be detected; the recognition result includes: an effect diagram corresponding to multiple processes.
  • the LOGO feature extracted above is input into a classifier trained based on a preset depth learning model, and after the classifier is identified, the final recognition result is obtained.
  • the recognition result is a plurality of processes such as thick plate glue, digital printing, appliqué embroidery, sequin embroidery, three-dimensional glue process, jacquard, three-dimensional thick plate glue, hollowing, patch seam and/or printing patch, etc. Effect chart.
  • the classifier based on the preset deep learning model training is obtained in the following manner, as shown in FIG. 2:
  • the clothing LOGO sample data includes LOGO features, matching degrees and recognition results.
  • the above machine learning algorithm may be a neighboring algorithm, a maximum expectation algorithm, and a support vector machine algorithm.
  • the specific algorithm may be selected according to a specific situation, which is not limited herein.
  • the clothing LOGO sample data includes triple data; wherein the triple data includes: source data, forward data belonging to the same category as the source data, and subordinate to the source data Reverse data for different categories.
  • the source data is sample data with the same recognition result randomly obtained from the clothing LOGO sample data.
  • the forward data is sample data that is randomly obtained from the clothing LOGO sample data and is consistent with the recognition result of the source data; the matching degree of the source data is higher than the matching degree of the forward data.
  • the reverse data is sample data that is randomly obtained from the clothing LOGO sample data and is inconsistent with the recognition result of the source data.
  • the ternary data is: a first picture with good performance of the clothing LOGO image in the clothing LOGO sample data, a second picture with poor performance of the clothing LOGO image in the clothing LOGO sample data, and the first picture
  • the picture and the second picture identify a third picture that is different in result.
  • the specific first picture is the source data with the highest matching degree, and the second picture has a gap with the first picture in terms of definition and/or resolution, and the matching degree is lower than the first picture, but still is the clothing LOGO image, Positive data.
  • the third picture is the reverse data of the reverse contrast in the training, and the positive opposition ratio is used to further enhance the recognition ability of the classifier, and improve the detection and recognition accuracy of the anti-neural network clothing LOGO effect.
  • the classifier for training based on the preset depth learning model is used to identify the LOGO feature, and the recognition result of the LOGO image to be detected is generated, which specifically includes the following steps, as shown in FIG. 3:
  • S301 Determine a data format corresponding to the classifier trained based on the preset depth learning model.
  • the classifier of the preset deep learning model training supports the binary data format
  • the LOGO feature is converted, converted into a binary data format, and then input into the trained classifier for recognition, because the machine language is Binary, so the binary data format can speed up the process of identification, and no additional data conversion is required when identifying, so that the efficiency of recognition can be improved.
  • the clothing LOGO recognition effect of the classifier trained based on the deep learning model is more accurate, so that the LOGO renderings under the various processes presented are more accurate.
  • the LOGO image of the garment to be inspected is first obtained, and then the LOGO feature of the LOGO image of the garment to be detected is extracted according to the preset depth learning model, wherein the preset depth learning model includes an anti-neural nerve.
  • the deep learning model of the network finally uses the classifier based on the preset deep learning model to identify the LOGO features, and generates the recognition result of the LOGO image to be detected.
  • the recognition result includes: a rendering diagram corresponding to various processes.
  • the clothing LOGO effect detection method can identify the clothing LOGO image through the deep learning model based on the anti-neural network, and generate the effect diagrams under different processes for the reference of the staff, without the need to proof the clothing LOGO, saving a large amount of Time and cost.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • the embodiment of the present application provides a garment LOGO effect detecting device.
  • the device includes an image acquiring module 51, a feature extracting module 52, and a clothing LOGO identifying module 53.
  • the image obtaining module 51 is configured to obtain a LOGO image to be detected.
  • the feature extraction module 52 is configured to extract a LOGO feature of the LOGO image to be detected according to the preset depth learning model, where the preset depth learning model includes an anti-neural network. a deep learning model;
  • the clothing LOGO identification module 53 is configured to identify the clothing LOGO feature by using a classifier based on the preset depth learning model training, and generate a recognition result of the LOGO image to be detected; the recognition result includes: a plurality of process corresponding effects Figure.
  • a variety of processes include: thick plate glue, digital printing, appliqué embroidery, sequin embroidery, three-dimensional plastic process, jacquard, three-dimensional thick plate glue, hollow, patch seam and / or printed patch.
  • Embodiment 3 is a diagrammatic representation of Embodiment 3
  • the embodiment of the present application further provides an electronic device.
  • the electronic device includes: a processor 60, a memory 61, a bus 62, and a communication interface 63.
  • the processor 60, the communication interface 63, and the memory 61 pass through the bus.
  • 62 is connected; the processor 60 is configured to execute an executable module, such as a computer program, stored in the memory 61.
  • the steps of the method as described in the method embodiments are implemented when the processor executes a computer program.
  • the memory 61 may include a high speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.
  • RAM high speed random access memory
  • non-volatile memory such as at least one disk memory.
  • the communication connection between the system network element and at least one other network element is implemented by at least one communication interface 63 (which may be wired or wireless), and may use an Internet, a wide area network, a local network, a metropolitan area network, or the like.
  • the bus 62 can be an ISA bus, a PCI bus, or an EISA bus.
  • the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one double-headed arrow is shown in Figure 6, but it does not mean that there is only one bus or one type of bus.
  • the memory 61 is configured to store a program, and the processor 60 executes the program after receiving the execution instruction, and the method executed by the device defined by the flow process disclosed in any embodiment of the present application may be applied to the processing.
  • processor 60 or implemented by processor 60.
  • Processor 60 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method may be completed by an integrated logic circuit of hardware in the processor 60 or an instruction in the form of software.
  • the processor 60 may be a general-purpose processor, including a central processing unit (CPU) and/or a network processor (NP), etc., or a digital signal processor (Digital Signal Processing, DSP), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device and/or discrete hardware Component.
  • DSP Central Processing unit
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • the methods, steps, and logical block diagrams disclosed in the embodiments of the present application can be implemented or executed.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the steps of the method disclosed in the embodiments of the present application may be directly implemented by the hardware decoding processor, or may be performed by a combination of hardware and software modules in the decoding processor.
  • the software modules can be located in conventional storage media such as random access memory, flash memory, read only memory, programmable read only memory, electrically erasable programmable memory, and/or registers.
  • the storage medium is located in the memory 61, and the processor 60 reads the information in the memory 61 and performs the steps of the above method in combination with its hardware.
  • the computer program product of the method for locating a network device includes a computer readable storage medium storing non-volatile program code executable by a processor, and the program code includes instructions for executing the foregoing method.
  • the program code includes instructions for executing the foregoing method.
  • each block of the flowchart or block diagram can represent a module, a program segment, or a portion of code that includes one or more of the Executable instructions.
  • the functions noted in the blocks may also occur in a different order than that illustrated in the drawings. For example, two consecutive blocks may be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of 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 function. Or it can be implemented by a combination of dedicated hardware and computer instructions.
  • the disclosed systems, devices, and methods may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some communication interface, device or unit, and may be electrical, mechanical or otherwise.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the functions, if implemented in the form of software functional units and sold or used as separate products, may be stored in a non-transitory computer readable storage medium executable by a processor.
  • the technical solution of the present application which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Multimedia (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)

Abstract

一种服装LOGO效果检测方法、装置及电子设备，涉及图像识别技术领域，其中，服装LOGO效果检测方法包括：获取待检测服装LOGO图像（S101）；根据预设深度学习模型提取待检测服装LOGO图像的LOGO特征，预设深度学习模型包括基于对抗神经网络的深度学习模型（S102）；利用基于预设深度学习模型训练的分类器对LOGO特征进行识别，生成待检测服装LOGO图像的识别结果；识别结果包括：多种工艺对应的效果图（S103）。该方法能够通过基于对抗神经网络的深度学习模型，对服装LOGO图像进行识别，生成不同工艺下的效果图，以供工作人员进行参考，不需要对服装LOGO进行打样，节省了大量时间和成本。

Description

服装LOGO效果检测方法、装置及电子设备
相关申请的交叉引用
本申请要求于2017年11月14日提交中国专利局的申请号为CN201711126619.7、名称为“服装LOGO效果检测方法、装置及电子设备”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。
技术领域
本申请涉及图像识别技术领域，尤其是涉及一种服装LOGO效果检测方法、装置及电子设备。
背景技术
随着我国经济的持续快速发展，服装业呈现蓬勃发展态势。近年来，服装LOGO的材质及样式越来越丰富，而且生产工艺多种多样，因而同样的LOGO设计在不同的工艺下的效果会有很大的差别，为了对比设计出来的服装LOGO在不同工艺下的效果，目前采用的方式是：针对设计出的服装LOGO，分别利用不同的工艺生产出样本，通过多种样本效果的对比，判断选择哪种工艺进行LOGO生产。此过程费时费力，且效率低下，尤其对于复杂工艺来说，其投入的成本很大。
发明内容
有鉴于此，本申请的目的在于提供一种服装LOGO效果检测方法、装置及电子设备，能够通过基于对抗神经网络的深度学习模型，对服装LOGO图像进行识别，生成不同工艺下的效果图，以供工作人员进行参考，不需要对服装LOGO进行打样，节省了大量时间和成本。
第一方面，本申请实施例提供了一种服装LOGO效果检测方法，包括：
获取待检测服装LOGO图像；
根据预设深度学习模型提取待检测服装LOGO图像的LOGO特征，预设深度学习模型包括基于对抗神经网络的深度学习模型；
利用基于预设深度学习模型训练的分类器对LOGO特征进行识别，生成待检测服装LOGO图像的识别结果；识别结果包括：多种工艺对应的效果图。
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，多种工艺包括：厚板胶、数码印、贴布绣、珠片绣、立体胶工艺、提花、立体厚板胶、镂空、贴布缝和/或印花贴布。
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，基于对抗神经网络的深度学习模型是通过数量超过一定阈值的服装LOGO样本数据训练得到 的，服装LOGO样本数据包括不同工艺的服装LOGO对应的图片。
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，通过以下方式获得基于预设深度学习模型训练的分类器：
利用基于对抗神经网络的深度学习模型提取服装LOGO样本数据的深层特征；
基于机器学习算法，对深层特征训练分类器；
其中服装LOGO样本数据中包括LOGO特征、匹配度和识别结果。
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，利用基于预设深度学习模型训练的分类器对LOGO特征进行识别，生成待检测服装LOGO图像的识别结果，具体包括：
确定基于预设深度学习模型训练的分类器所对应的数据格式；
如果数据格式包括二进制数据格式，将LOGO特征转换为二进制数据格式后输入基于预设深度学习模型训练的分类器，以生成对应待检测服装LOGO图像的识别结果。
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，在利用基于预设深度学习模型训练的分类器对LOGO特征进行识别，生成待检测服装LOGO图像的识别结果之后，还包括：
将待检测服装LOGO图像的识别结果作为新的服装LOGO样本数据保存在服装LOGO样本数据库中。
第二方面，本申请实施例提供一种服装LOGO效果检测装置，装置包括：
图像获取模块，配置成获取待检测服装LOGO图像；
特征提取模块，配置成根据预设深度学习模型提取待检测服装LOGO图像的LOGO特征，预设深度学习模型包括基于对抗神经网络的深度学习模型；
服装LOGO识别模块，配置成利用基于预设深度学习模型训练的分类器对LOGO特征进行识别，生成待检测服装LOGO图像的识别结果；识别结果包括：多种工艺对应的效果图。
结合第二方面，本申请实施例提供了第二方面的第一种可能的实施方式，其中，多种工艺包括：厚板胶、数码印、贴布绣、珠片绣、立体胶工艺、提花、立体厚板胶、镂空、贴布缝和/或印花贴布。
第三方面，本申请实施例提供一种电子设备，包括存储器和处理器，存储器上存储有可在处理器上运行的计算机程序，处理器执行计算机程序时实现第一方面所述的方法的步骤。
第四方面，本申请实施例还提供一种具有处理器可执行的非易失的程序代码的计算机可读介质，程序代码使处理器执行如第一方面所述的方法。
本申请实施例带来了以下有益效果：
在本申请实施例提供的服装LOGO效果检测方法中，首先获取待检测服装LOGO图像；然后根据预设深度学习模型提取待检测服装LOGO图像的LOGO特征，其中，预设深度学习模型包括基于对抗神经网络的深度学习模型；最后利用基于预设深度学习模型训练的分类器对LOGO特征进行识别，生成待检测服装LOGO图像的识别结果；识别结果包括：多种工艺对应的效果图。该方法能够通过基于对抗神经网络的深度学习模型，对服装LOGO图像进行识别，生成不同工艺下的效果图，以供工作人员进行参考，不需要对服装LOGO进行打样，节省了大量时间和成本。
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。
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。
附图说明
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。
图1为本申请实施例一提供的一种服装LOGO效果检测方法的流程图；
图2为本申请实施例一提供的另一种服装LOGO效果检测方法的流程图；
图3为本申请实施例一提供的另一种服装LOGO效果检测方法的流程图；
图4为本申请实施例一提供的另一种服装LOGO效果检测方法的流程图；
图5为本申请实施例二提供的一种服装LOGO效果检测装置的结构示意图；
图6为本申请实施例三提供的一种电子设备的结构示意图。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合附图对本申请的技术方案进行清楚且完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动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。
目前现有的服装LOGO效果检测方式，需要针对同一个LOGO，生产出多种工艺下的样本，这种方式不仅费时费力，而且效率低下，尤其对于复杂工艺来说，其花费的时间和成本都很大。基于此，本申请实施例提供的服装LOGO效果检测方法、装置及电子设备， 能够通过基于对抗神经网络的深度学习模型，对服装LOGO图像进行识别，生成不同工艺下的效果图，以供工作人员进行参考，不需要对服装LOGO进行打样，节省了大量时间和成本。
为便于对本实施例进行理解，首先对本申请实施例所公开的一种服装LOGO效果检测方法进行详细介绍。
实施例一：
本申请实施例提供了一种服装LOGO效果检测方法，该方法可以应用于服装生产领域。参见图1所示，该方法包括：
S101：获取待检测服装LOGO图像。
具体的，通过摄像机对设计出来的服装LOGO进行拍照，从而采集到待检测服装LOGO图像。
S102：根据预设深度学习模型提取待检测服装LOGO图像的LOGO特征，预设深度学习模型包括基于对抗神经网络的深度学习模型。
基于对抗神经网络的深度学习模型是通过数量超过一定阈值的服装LOGO样本数据训练得到的，服装LOGO样本数据包括利用不同工艺生产出的服装LOGO对应的图片。
在一个优选的实施方式中，上述多种基于对抗神经网络的深度学习模型可以通过Caffe深度学习框架实现。
具体地，服装LOGO样本数据中的服装LOGO图像的数量越多越好，工艺种类越多越好，比如厚板胶、数码印、贴布绣、珠片绣、立体胶工艺、提花、立体厚板胶、镂空、贴布缝和/或印花贴布等。数据越多，训练生成的基于对抗神经网络的深度学习模型的通用性越好。这样有利于后续对待检测服装LOGO图像的准确识别，并呈现出多种工艺下的效果图，为相关工作人员提供LOGO效果参考。
该步骤S102具体包括：将待检测服装LOGO图像作为输入图像在预设深度学习模型中包含的多个基层中依次进行特征训练，当训练完成后，提取多个集成中的全连接层或者其他指定基层输出的特征向量作为该待检测服装LOGO图像中对应的LOGO特征。
S103：利用基于预设深度学习模型训练的分类器对LOGO特征进行识别，生成待检测服装LOGO图像的识别结果；所述识别结果包括：多种工艺对应的效果图。
将上述提取的LOGO特征输入基于预设深度学习模型训练的分类器，通过该分类器识别后，获得最终的识别结果。具体地，识别结果为多种工艺如厚板胶、数码印、贴布绣、珠片绣、立体胶工艺、提花、立体厚板胶、镂空、贴布缝和/或印花贴布等对应的效果图。
在一个可选的实施方式中，通过以下方式获得基于预设深度学习模型训练的分类器， 参见图2所示：
S201：利用基于对抗神经网络的深度学习模型提取服装LOGO样本数据的深层特征。
S202：基于机器学习算法，对深层特征训练分类器。
其中服装LOGO样本数据中包括LOGO特征、匹配度和识别结果。上述机器学习算法可以是邻近算法、最大期望算法及支持向量机算法等，具体算法可以根据具体情况选择，这里不作限定。
在一个可选的实施例中，上述服装LOGO样品数据包括三元组数据；其中该三元组数据包括：源数据、与所述源数据属于同一类别的正向数据以及与该源数据分属不同类别的反向数据。
其中，源数据为从服装LOGO样品数据中随机获取到的识别结果相同的样本数据。正向数据为从服装LOGO样本数据中随机获取的与源数据的识别结果一致的样本数据；该源数据的匹配度高于正向数据的匹配度。反向数据为从服装LOGO样本数据中随机获取的与源数据的识别结果不一致的样本数据。
在一个具体的实施方式中，三元组数据分别为：服装LOGO样本数据中服装LOGO图像性能良好的第一图片，服装LOGO样本数据中服装LOGO图像性能较差的第二图片，以及与第一图片和第二图片识别结果不同的第三图片。具体的第一图片为匹配程度最高的源数据，第二图片在清晰度和/或分辨率等方面与第一图片存在差距，其匹配度低于第一图片，但仍是服装LOGO图像，为正向数据。第三图片则是在训练是进行反向对比的反向数据，以次通过正反对比，进一步增强了分类器的识别能力，提高了对抗神经网络服装LOGO效果检测识别准确性。
利用基于预设深度学习模型训练的分类器对LOGO特征进行识别，生成待检测服装LOGO图像的识别结果，具体包括以下步骤，参见图3所示：
S301：确定基于预设深度学习模型训练的分类器所对应的数据格式。
S302：如果数据格式包括二进制数据格式，将LOGO特征转换为二进制数据格式后输入基于预设深度学习模型训练的分类器，以生成对应待检测服装LOGO图像的识别结果。
具体的，若预设深度学习模型训练的分类器支持二进制数据格式，那么就将LOGO特征进行转换，转换为二进制的数据格式，再输入到上述训练的分类器中，进行识别，由于机器语言为二进制，因此通过二进制的数据格式，可以加快识别的过程，在进行识别时，不需要再进行额外的数据转换，国此可以提高识别的效率。
此外，在生成待检测服装LOGO图像的识别结果之后，还包括以下步骤，参见图4所示：
S401：将待检测服装LOGO图像的识别结果作为新的服装LOGO样本数据保存在服装 LOGO样本数据库中。
通过不断地更新服装LOGO样本数据库中的数据，使得基于深度学习模型训练出的分类器的服装LOGO识别效果更加精确，以使呈现出的多种工艺下的LOGO效果图更加精确。
在本申请实施例提供的服装LOGO效果检测方法中，首先获取待检测服装LOGO图像，然后根据预设深度学习模型提取待检测服装LOGO图像的LOGO特征，其中，预设深度学习模型包括基于对抗神经网络的深度学习模型，最后利用基于预设深度学习模型训练的分类器对LOGO特征进行识别，生成待检测服装LOGO图像的识别结果，识别结果包括：多种工艺对应的效果图。该服装LOGO效果检测方法能够通过基于对抗神经网络的深度学习模型，对服装LOGO图像进行识别，生成不同工艺下的效果图，以供工作人员进行参考，不需要对服装LOGO进行打样，节省了大量时间和成本。
实施例二：
本申请实施例提供一种服装LOGO效果检测装置，参见图5所示，该装置包括：图像获取模块51、特征提取模块52和服装LOGO识别模块53。
其中，图像获取模块51，配置成获取待检测服装LOGO图像；特征提取模块52，配置成根据预设深度学习模型提取待检测服装LOGO图像的LOGO特征，预设深度学习模型包括基于对抗神经网络的深度学习模型；服装LOGO识别模块53，配置成利用基于预设深度学习模型训练的分类器对服装LOGO特征进行识别，生成待检测服装LOGO图像的识别结果；识别结果包括：多种工艺对应的效果图。其中，多种工艺包括：厚板胶、数码印、贴布绣、珠片绣、立体胶工艺、提花、立体厚板胶、镂空、贴布缝和/或印花贴布。
本申请实施例所提供的服装LOGO效果检测装置中，各个模块的工作过程与前述服装LOGO效果检测方法具有相同的技术特征，因此，同样可以实现上述功能，在此不再赘述。
实施例三：
本申请实施例还提供一种电子设备，参见图6所示，该电子设备包括：处理器60，存储器61，总线62和通信接口63，所述处理器60、通信接口63和存储器61通过总线62连接；处理器60配置成执行存储器61中存储的可执行模块，例如计算机程序。处理器执行计算机程序时实现如方法实施例所述的方法的步骤。
其中，存储器61可能包含高速随机存取存储器(RAM，Random Access Memory)，也可能还包括非不稳定的存储器(non-volatile memory)，例如至少一个磁盘存储器。通过至少一个通信接口63(可以是有线或者无线)实现该系统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。
总线62可以是ISA总线、PCI总线或EISA总线等。所述总线可以分为地址总线、数 据总线和控制总线等。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。
其中，存储器61配置成存储程序，所述处理器60在接收到执行指令后，执行所述程序，前述本申请实施例任一实施例揭示的流过程定义的装置所执行的方法可以应用于处理器60中，或者由处理器60实现。
处理器60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60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60可以是通用处理器，包括中央处理器(Central Processing Unit，简称CPU)和/或网络处理器(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和/或分立硬件组件。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、电可擦写可编程存储器和/或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61，处理器60读取存储器61中的信息，结合其硬件完成上述方法的步骤。
本申请实施例所提供的网络设备的定位方法的计算机程序产品，包括存储了处理器可执行的非易失的程序代码的计算机可读存储介质，所述程序代码包括的指令可用于执行前面方法实施例中所述的方法，具体实现可参见方法实施例，在此不再赘述。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的装置及电子设备的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
附图中的流程图和框图显示了根据本申请的多个实施例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系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
在本申请的描述中，需要说明的是，术语“中心”、“上”、“下”、“左”、“右”、“竖直”、“水平”、“内”或“外”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本申请和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位或者以特定的方位构造和操作，因此不能理解为对本申请的限制。此外，术语“第一”、“第二”或“第三”仅用于描述目的，而不能理解为指示或暗示相对重要性。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，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，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，又例如，多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些通信接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，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单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个处理器可执行的非易失的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。
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。

Claims (10)

  1. 一种服装LOGO效果检测方法，其特征在于，包括：
    获取待检测服装LOGO图像；
    根据预设深度学习模型提取所述待检测服装LOGO图像的LOGO特征，所述预设深度学习模型包括基于对抗神经网络的深度学习模型；
    利用基于所述预设深度学习模型训练的分类器对所述LOGO特征进行识别，生成所述待检测服装LOGO图像的识别结果；所述识别结果包括：多种工艺对应的效果图。
  2. 根据权利要求1所述的方法，其特征在于，所述多种工艺包括：厚板胶、数码印、贴布绣、珠片绣、立体胶工艺、提花、立体厚板胶、镂空、贴布缝和/或印花贴布。
  3. 根据权利要求1所述的方法，其特征在于，所述基于对抗神经网络的深度学习模型是通过数量超过一定阈值的服装LOGO样本数据训练得到的，所述服装LOGO样本数据包括不同工艺的服装LOGO对应的图片。
  4. 根据权利要求1所述的方法，其特征在于，通过以下方式获得所述基于所述预设深度学习模型训练的分类器：
    利用基于对抗神经网络的深度学习模型提取服装LOGO样本数据的深层特征；
    基于机器学习算法，对所述深层特征训练分类器；
    其中所述服装LOGO样本数据中包括LOGO特征、匹配度和识别结果。
  5. 根据权利要求1所述的方法，其特征在于，所述利用基于所述预设深度学习模型训练的分类器对所述LOGO特征进行识别，生成所述待检测服装LOGO图像的识别结果，具体包括：
    确定所述基于所述预设深度学习模型训练的分类器所对应的数据格式；
    如果所述数据格式包括二进制数据格式，将所述LOGO特征转换为二进制数据格式后输入所述基于所述预设深度学习模型训练的分类器，以生成对应所述待检测服装LOGO图像的识别结果。
  6. 根据权利要求1-5任一项所述的方法，其特征在于，在所述利用基于所述预设深度学习模型训练的分类器对所述LOGO特征进行识别，生成所述待检测服装LOGO图像的识别结果之后，还包括：
    将所述待检测服装LOGO图像的识别结果作为新的服装LOGO样本数据保存在服装LOGO样本数据库中。
  7. 一种服装LOGO效果检测装置，其特征在于，所述装置包括：
    图像获取模块，配置成获取待检测服装LOGO图像；
    特征提取模块，配置成根据预设深度学习模型提取所述待检测服装LOGO图像的LOGO特征，所述预设深度学习模型包括基于对抗神经网络的深度学习模型；
    服装LOGO识别模块，配置成利用基于所述预设深度学习模型训练的分类器对所述LOGO特征进行识别，生成所述待检测服装LOGO图像的识别结果；所述识别结果包括：多种工艺对应的效果图。
  8. 根据权利要求7所述的装置，其特征在于，所述多种工艺包括：厚板胶、数码印、贴布绣、珠片绣、立体胶工艺、提花、立体厚板胶、镂空、贴布缝和/或印花贴布。
  9. 一种电子设备，包括存储器和处理器，所述存储器上存储有可在所述处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现上述权利要求1至7任一项所述的方法的步骤。
  10. 一种具有处理器可执行的非易失的程序代码的计算机可读介质，其特征在于，所述程序代码使所述处理器执行所述权利要求1至7任一项所述的方法。
PCT/CN2018/115496 2017-11-14 2018-11-14 服装logo效果检测方法、装置及电子设备 WO2019096179A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711126619.7A CN107845092A (zh) 2017-11-14 2017-11-14 服装logo效果检测方法、装置及电子设备
CN201711126619.7 2017-11-14

Publications (1)

Publication Number Publication Date
WO2019096179A1 true WO2019096179A1 (zh) 2019-05-23

Family

ID=61679478

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/115496 WO2019096179A1 (zh) 2017-11-14 2018-11-14 服装logo效果检测方法、装置及电子设备

Country Status (2)

Country Link
CN (1) CN107845092A (zh)
WO (1) WO2019096179A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111161295A (zh) * 2019-12-30 2020-05-15 神思电子技术股份有限公司 一种菜品图像背景剥离方法
CN112150440A (zh) * 2020-09-23 2020-12-29 创新奇智(青岛)科技有限公司 一种成衣质检算法的测试方法和装置
CN112734383A (zh) * 2021-01-12 2021-04-30 浙江理工大学 一种自动识别服装工艺的方法

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107845092A (zh) * 2017-11-14 2018-03-27 深圳码隆科技有限公司 服装logo效果检测方法、装置及电子设备
CN110147854A (zh) * 2018-07-19 2019-08-20 深圳码隆科技有限公司 服装识别方法、计算机设备和存储介质
CN109034255A (zh) * 2018-08-02 2018-12-18 深圳码隆科技有限公司 服装销量的预测方法及装置
CN109190671A (zh) * 2018-08-02 2019-01-11 深圳码隆科技有限公司 服装关键点测量方法及装置
CN110941188A (zh) * 2018-09-25 2020-03-31 珠海格力电器股份有限公司 智能家居控制方法及装置
CN109670591B (zh) * 2018-12-14 2022-09-27 深圳市商汤科技有限公司 一种神经网络的训练方法及图像匹配方法、装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101782771A (zh) * 2010-03-17 2010-07-21 东华大学 基于免疫神经网络的纺丝工艺智能优化设计方法
CN105631461A (zh) * 2016-02-01 2016-06-01 信阳农林学院 一种图像识别系统及方法
US20160342859A1 (en) * 2015-05-18 2016-11-24 Facebook, Inc. Logo detection
CN107204025A (zh) * 2017-04-18 2017-09-26 华北电力大学 基于视觉感知的自适应服装动画建模方法
US20170316281A1 (en) * 2016-04-28 2017-11-02 Microsoft Technology Licensing, Llc Neural network image classifier
CN107845092A (zh) * 2017-11-14 2018-03-27 深圳码隆科技有限公司 服装logo效果检测方法、装置及电子设备

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9665722B2 (en) * 2012-08-10 2017-05-30 Visa International Service Association Privacy firewall
CN107292813B (zh) * 2017-05-17 2019-10-22 浙江大学 一种基于生成对抗网络的多姿态人脸生成方法
CN107220929A (zh) * 2017-06-23 2017-09-29 深圳市唯特视科技有限公司 一种使用循环一致对抗网络的非成对图像转化方法
CN107273936B (zh) * 2017-07-07 2020-09-11 广东工业大学 一种gan图像处理方法及系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101782771A (zh) * 2010-03-17 2010-07-21 东华大学 基于免疫神经网络的纺丝工艺智能优化设计方法
US20160342859A1 (en) * 2015-05-18 2016-11-24 Facebook, Inc. Logo detection
CN105631461A (zh) * 2016-02-01 2016-06-01 信阳农林学院 一种图像识别系统及方法
US20170316281A1 (en) * 2016-04-28 2017-11-02 Microsoft Technology Licensing, Llc Neural network image classifier
CN107204025A (zh) * 2017-04-18 2017-09-26 华北电力大学 基于视觉感知的自适应服装动画建模方法
CN107845092A (zh) * 2017-11-14 2018-03-27 深圳码隆科技有限公司 服装logo效果检测方法、装置及电子设备

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111161295A (zh) * 2019-12-30 2020-05-15 神思电子技术股份有限公司 一种菜品图像背景剥离方法
CN111161295B (zh) * 2019-12-30 2023-11-21 神思电子技术股份有限公司 一种菜品图像背景剥离方法
CN112150440A (zh) * 2020-09-23 2020-12-29 创新奇智(青岛)科技有限公司 一种成衣质检算法的测试方法和装置
CN112734383A (zh) * 2021-01-12 2021-04-30 浙江理工大学 一种自动识别服装工艺的方法
CN112734383B (zh) * 2021-01-12 2024-04-12 浙江理工大学 一种自动识别服装工艺的方法

Also Published As

Publication number Publication date
CN107845092A (zh) 2018-03-27

Similar Documents

Publication Publication Date Title
WO2019096179A1 (zh) 服装logo效果检测方法、装置及电子设备
WO2019096178A1 (zh) 纤维检测方法、装置及电子设备
CN109947967B (zh) 图像识别方法、装置、存储介质和计算机设备
US11741593B2 (en) Product defect detection method, device and system
WO2019096174A1 (zh) 人物图像服装颜色识别方法、装置及电子设备
CN109034159A (zh) 图像信息提取方法和装置
CN111832447B (zh) 建筑图纸构件识别方法、电子设备及相关产品
CN104899060B (zh) 一种图片加载处理方法和装置
WO2016187888A1 (zh) 基于字符识别的关键词通知方法及设备、计算机程序产品
CN107845113A (zh) 目标元素定位方法、装置以及用户界面测试方法、装置
CN110232379A (zh) 一种车辆姿态检测方法及系统
WO2019105457A1 (zh) 图像处理方法、计算机设备和计算机可读存储介质
CN114862845B (zh) 手机触摸屏的缺陷检测方法、装置、设备及存储介质
CN109348731A (zh) 一种图像匹配的方法及装置
CN105589801B (zh) 手机集群测试方法和系统
TWI557698B (zh) 用於取得地圖的方法、電子裝置和電腦可讀媒體
WO2015172359A1 (zh) 一种对象搜索方法及装置
TW201317904A (zh) 標籤檢測系統、裝置及其檢測標籤的方法
WO2019090904A1 (zh) 确定距离的方法、装置、设备及存储介质
US20220207724A1 (en) Method of determining a distribution of stem cells in a cell image, electronic device, and storage medium
TWI671686B (zh) 影像數據擷取方法及影像數據擷取裝置
CN110443328A (zh) 基于led防伪标签的识别验伪方法、装置、设备及介质
CN115546219B (zh) 检测板式生成方法、板卡缺陷检测方法、装置及产品
JP2011258036A (ja) ３次元形状検索装置、３次元形状検索方法、及びプログラム
CN111914863A (zh) 目标检测方法、装置、终端设备及计算机可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18877810

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18877810

Country of ref document: EP

Kind code of ref document: A1