A Target Recognition Method Based on Statistical Matching of Local Similar Structures
A target recognition and local similarity technology, applied in the field of image processing, can solve the problems of difficult to achieve accurate recognition and detection, poor recognition accuracy, missed detection, etc., to achieve the effect of reducing the amount of calculation, high recognition accuracy, and saving the training process
Active Publication Date: 2018-09-04
NANJING UNIV OF SCI & TECH
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, using an example picture to recognize the test picture is only suitable for targets with compact overall structure and simple pose changes, and the detection of specific poses has a good recognition effect. For targets with large changes in overall characteristics or with pose diversity , when there is a big difference between the target pose in the example picture and the test picture, it is difficult to achieve the purpose of accurate recognition and detection by using another example picture, and there will be missed detection, and the recognition accuracy is poor
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
[0015] Basic train of thought of the present invention:
[0016] For the target in the test picture to be recognized, since the specific pose of the target in the test picture is not known in advance, multiple pictures containing the target in different shapes and scales are selected to form an example picture set. The present invention judges whether there is a target similar to the sample image in the region according to the number of similar structures contained in a certain region in the test picture, because the local adaptive kernel regression method (LARK) is in the extracted local structure feature It has good robustness, and the obtained structural features can well describe the local structural features of the image, so similar structural features can be expressed as local structural similarities.
[0017] The inventive method comprises the following steps:
[0018] Step 1. For example picture set Q={Q 1 ,Q 2 ,...,Q n} of n pictures and the size is m 1 × m 2 Th...
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 a target recognition method based on collecting and matching a local similar structure. The method comprises: firstly, establishing an example picture set including different attitudes and scales of a target; extracting local structural features to obtain a local structural feature matrix of the example picture set; simplifying the structural feature matrix by removing similar structural features; then, obtaining a similarity image of a tested picture and the example picture set through a matching method of collecting the number of structural properties, similar to the structural feature matrix, in the local area of the tested picture; and finally extracting the target from the similarity image by using a method of non-maximum suppression to achieve the purpose of target recognition. The method in the invention can effectively recognize the target with complex structure and attitude diversity.
Description
technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to an object recognition method based on statistical matching of local similar structures. Background technique [0002] Object recognition is an important field of computer artificial intelligence. In current research, the more popular theoretical models of object recognition are based on probability models and part and shape models. These recognition methods are implemented on the basis of statistical learning-based classifiers, and require a large amount of learning and training to determine the parameters of each classifier included, which is also called a parameter method. Generally, this method requires a large number of training samples in the learning process, but the training process will lead to overfitting of the training parameters, and the training process is very slow. In order to avoid using a large number of samples and a long training process i...
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): G06K9/64
Inventor 柏连发张毅罗飞扬韩静祁伟陈钱顾国华
Owner NANJING UNIV OF SCI & TECH