Image comparison method and image comparison apparatus
A technology of matching method and image comparison, which is applied in the fields of image comparison method and image comparison device, and can solve problems such as unsatisfactory image retrieval results, large investment in manpower and material resources, and low efficiency
Inactive Publication Date: 2017-01-04
SHENZHEN SENSETIME TECH CO LTD
View PDF4 Cites 10 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, this method has a huge investment in manpower and material resources, and the efficiency is not high, and it cannot meet the timely requirements of the criminal investigation department.
Secondly, image retrieval and analysis is performed by using equipment in the image search method, but due to the low image feature extraction ability of traditional algorithms, especially poor adaptability to blurred images, dark light, noise, multi-angle, occlusion, etc., the image retrieval results are not good. ideal
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0060] In one embodiment, an image comparison method is disclosed, such as figure 1 As shown, the method includes the following steps:
[0061] S100. Use the deep convolutional neural network model to extract the biological identity features of each image to be compared;
[0062] S200: According to the degree of similarity between any two biological identity features, compare the images to be compared corresponding to the any two biological identity features.
[0063] In this embodiment, the method uses the deep convolutional neural network model to effectively extract the biometric characteristics in the image, which is beneficial to improve the accuracy of identity recognition. Compared with the traditional image comparison method, the deep convolutional neural network The biological identity features learned by the network in the large-scale general-purpose face database have stronger identity representation ability and discrimination, which can significantly improve the identity ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention relates to an image comparison method and an image comparison apparatus. The method comprises the following steps that: the biometric identity features of each image to be compared are extracted through using a deep convolution neural network model; and images to be compared corresponding to any two biometric identity features are compared according to the similarity degree between the any two biometric identity features. According to the image comparison method and the image comparison apparatus of the invention, the accuracy of face recognition is effectively improved through one kind of identity characterization, and the trained deep convolution neural network model has a powerful and accurate biometric identity feature extraction ability.
Description
Technical field [0001] The present disclosure relates to video surveillance, and in particular, to an image comparison method and image comparison device. Background technique [0002] Image investigation is video image investigation. It refers to the discovery of investigation clues through the retrieval, viewing, analysis and research of surveillance images on the scene and surrounding areas according to the needs of the case, combining modern scientific and technological means with traditional investigation measures to form a unified A method system that involves multi-party participation and overall combat to achieve the purpose of combating, preventing and controlling criminal crimes is an important part of obtaining evidence and clues in the process of investigating and handling cases. [0003] There are many reasons why it is difficult for human eyes or traditional analysis equipment to accurately identify surveillance images. Among them, the most important factor is the qu...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V40/168G06V40/172G06F18/214
Inventor 贾思博陈治华陈宇恒王晶林俊郯张广程
Owner SHENZHEN SENSETIME TECH CO LTD