Object feature extraction method, device, storage medium and electronic equipment
A technology of object features and extraction methods, applied in character and pattern recognition, instruments, calculations, etc., can solve problems affecting accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0041] figure 1 is a flow chart showing the object feature extraction method according to Embodiment 1 of the present invention.
[0042] refer to figure 1 , in step S110, acquire feature data of multiple scales of the target object from the image to be checked.
[0043] Here, the image to be checked may be a static image or a video frame image containing the target object. Target objects can be objects with visible shapes such as pedestrians, vehicles, animals, flying objects, etc.
[0044] Feature data of multiple scales of the target object can be extracted from the image to be checked by an applicable image processing method. Here, multiple scales can be understood as multiple scale resolutions, and the feature data of any scale can be, for example, the texture feature data of the image, the color feature data of the image, the shape feature data of the object, or, from another perspective, for the The feature vector matrix of the scale-extracted image, etc.
[0045] ...
Embodiment 2
[0052] figure 2 It is a flow chart showing the object feature extraction method according to the second embodiment of the present invention.
[0053] For ease of description, in this disclosure, it is assumed that the aforementioned multiple scales are N scales, and N is an integer greater than 1.
[0054] In addition, according to this embodiment, the feature data of any scale includes the first feature map corresponding to multiple feature channels of the scale. Here, multiple feature channels may correspond to predetermined multiple image features or object features, so as to characterize the response degree of the image to the image features or object features corresponding to each feature channel.
[0055] refer to figure 2 , in step S210, obtain the first feature map of N scales of the target object from the image to be checked.
[0056] For example, multiple convolutions and multiple downsampling pooling are performed on the image to be checked to obtain the first ...
Embodiment 3
[0080] image 3 It is a flow chart showing the object feature extraction method according to the third embodiment of the present invention.
[0081] In the object feature extraction method according to Embodiment 3 of the present invention, the first neural network for generating fusion feature data and / or the second neural network for feature extraction can be used to perform the corresponding steps.
[0082] refer to image 3 , in step S310, through the second neural network, obtain the feature data of multiple scales of the target object from the image to be checked.
[0083] Specifically, the feature data of multiple scales of the target object can be obtained from the image to be checked through the pre-trained second neural network. The second neural network can obtain multi-scale feature data of the target object by performing multiple convolutions and pooling on the image to be checked.
[0084] As mentioned above, optionally, the feature data of any scale includes ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More