Image recognition method and system in bad weather
A technology of severe weather and recognition methods, applied in the direction of neural learning methods, character and pattern recognition, TV system components, etc., can solve problems such as high cost and high power consumption, and achieve cost reduction and power consumption, and clear division of labor Effect
Active Publication Date: 2020-11-20
UNIV OF SCI & TECH BEIJING
View PDF8 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The technical problem to be solved by the present invention is to provide an imaging recognition method and system under severe weather, so as to solve the problems of high cost and high power consumption in the prior art for target recognition on the PC side
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 1
[0051] Such as figure 1 As shown, the imaging recognition method under severe weather provided by the embodiment of the present invention includes:
[0052] S101, performing image enhancement processing on the collected video through a DSP (Digital Signal Processor) module, wherein DSP is digital signal processing;
[0053] S102, in the GPU (Graphics Processing Unit) module, build the same deep learning network as the computer side, and obtain the parameters of the deep learning network that has been trained on the computer side, wherein the GPU is an image processing unit, and the parameters include: weights value and offset value;
[0054] S103, the deep learning network in the GPU module extracts features of the enhanced image according to the parameters of the deep learning network trained on the computer, and performs image recognition and classification based on the extracted features.
[0055]In the imaging recognition method under bad weather described in the embodim...
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 invention provides an imaging identification method and system in severe weather. an image enhancement algorithm and a deep learning network are deployed in different processors in the same system, and the cost and power consumption can be reduced. The method comprises: image enhancement processing is conducted on an acquired video through a DSP module, and the DSP module is used for digital signal processing; in the GPU module, a deep learning network which is the same as the computer end is built, parameters of the deep learning network which is trained by the computer end are obtained,and the GPU is an image processing unit; and the deep learning network in the GPU module extracts the characteristics of the enhanced image according to the parameters of the deep learning network trained by the computer end, and identifies and classifies the image based on the extracted characteristics. The invention relates to the field of target classification and recognition.
Description
technical field [0001] The invention relates to the field of target classification and recognition, in particular to an imaging recognition method and system in bad weather. Background technique [0002] All images in nature are continuously changing simulated images. In some scenes in daily life, there are often more than one type and number of targets, such as: pedestrians, motor vehicles, and non-motor vehicles at intersections. Target classification and recognition has always been a hot research direction in computer vision and digital image processing, and is widely used in many fields such as intelligent transportation, intelligent video surveillance, military strike, and aerospace. Its purpose is how to accurately detect the target in the video or image to be detected in real time, and obtain the category of the target and the specific position in the image. [0003] The existing deep learning network for object recognition has a complex structure and a large network...
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): G06K9/00G06K9/32G06K9/34G06K9/54G06K9/62G06N3/04G06N3/08H04N5/217H04N9/64
Inventor 蓝金辉李建勇
Owner UNIV OF SCI & TECH BEIJING