Image recognition and neural network model training method, device and system
A neural network model and image technology, applied in the field of image processing, can solve the problem of high false recognition rate, and achieve the effect of reducing false recognition rate and improving adaptability and accuracy.
Active Publication Date: 2021-10-15
MEGVII BEIJINGTECH CO LTD
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] However, the above-mentioned cross-modal image comparison scheme has the problem of high false recognition rate
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
[0054] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not intended to limit the present application.
[0055] The image recognition method provided by this application can be applied, but not limited to, such as figure 1 shown in the application environment. Wherein, the shooting device 12 can acquire the image to be recognized of the object to be recognized, and send the image to be recognized to the computer device 11; the computer device 11 can extract the target image features from the image to be recognized, and send the image to be recognized The image features are compared with the image features of the bottom library image in the bottom library image group, and the...
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 application relates to an image recognition and neural network model training method, device, system and readable storage medium. The method includes: obtaining an image to be recognized; inputting the image to be recognized into a neural network model for feature extraction processing, and outputting target image features of the image to be recognized; the neural network model includes at least one space transformation block, and the space transformation block is used to adopt a space transformation matrix Perform space transformation processing on the input intermediate image features; the intermediate image features are generated during the feature extraction process of the image to be recognized; the target image features of the image to be recognized and the bottom image features of the bottom library image in the bottom library image group Compare to obtain the comparison result; wherein, the bottom library image group includes at least one bottom library image, the bottom library image and the image to be recognized are two images of different modalities, and the bottom library image feature is that the neural network model learns from the bottom library image extracted. The method improves the accuracy of cross-modal image comparison.
Description
technical field [0001] The present application relates to the technical field of image processing, in particular to an image recognition and neural network model training method, device, system and readable storage medium. Background technique [0002] At present, there are more and more application scenarios based on face recognition technology. Under normal circumstances, the base library image used for face recognition is an RGB image, and because of the limitations of the dark light environment and other conditions, the captured image is an IR (Infrad, infrared) image, so there are face comparisons such as RGB images and IR images. Such cross-modal image comparison needs. [0003] A traditional solution is to train a convolutional neural network through the cross-entropy loss of multi-modal image training samples, and realize cross-modal image comparison and recognition based on the trained convolutional neural network. [0004] However, the above-mentioned cross-modal...
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 Patents(China)
IPC IPC(8): G06K9/00G06N3/08
CPCG06N3/08G06V40/168G06V40/172
Inventor 杜佳慧
Owner MEGVII BEIJINGTECH CO LTD