Color image description method based on color-dependent local binary pattern
A local binary mode and color image technology, applied in image enhancement, image analysis, image data processing, etc., can solve the problem of not having a good balance between length and feature performance, so as to reduce the amount of classification calculation and avoid the curse of dimensionality , The effect of speeding up the classification speed
Pending Publication Date: 2021-03-12
CHONGQING UNIV OF POSTS & TELECOMM
View PDF9 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Although these methods have certain effects on color image recognition, they do not balance the relationship between the length of the color image feature vector and the feature performance. In addition, how to better combine the color information of the image is also a consideration. The key issue
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] Below in conjunction with accompanying drawing, the present invention will be further described:
[0016] Such as figure 1 As shown, an image description method based on color-related local binary patterns includes the following steps:
[0017] The first step: first preprocess the color image data to obtain three relative similarity space (RSS) component images and three RGB color space component images of the color image;
[0018] The second step: using the multi-channel coding local binary pattern method (mdLBP) to perform feature extraction on the component image obtained in the first step, and obtain the mdLBP feature vector of the image data set;
[0019] The third step: determine the number of dominant modes: select the mdLBP feature vector of a part of the image randomly in the class for feature learning;
[0020] Step 4: According to the dominant mode table learned in the third step, perform dimension selection on the mdLBP feature vectors of all images in 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 invention discloses a color image description method based on a color correlation local binary pattern, and relates to the field of digital image processing. The method comprises the following steps: firstly, preprocessing a color image in a data set to obtain three component images of the color image in an RSS space and three component images of the color image in an RGB space; performing feature extraction on the component image of the color image by adopting a multi-channel coding local binary pattern method (mdLBP) to obtain an mdLBP feature vector of the image; selecting mdLBP featurevectors of a part of images in an intra-class random mode to perform feature learning, performing dimension selection on the mdLBP feature vectors of all the images in the data set according to the learned dominant mode table, and combining the selected features into a color-related local binary mode feature vector for describing the images; and finally, matching the test set with the training set to obtain a classification result. According to the method, the related information of different component images of the color image and the dominant mode of the LBP feature vector are utilized, sothat the accuracy of image description is effectively improved.
Description
technical field [0001] The invention belongs to the technical fields of digital image processing and machine vision. Specifically, it is a color image description method based on color-correlated local binary patterns. Background technique [0002] Image feature extraction and description is one of the challenging research topics in the field of image processing and pattern recognition, and at the same time, it has increasingly become a part closely related to life. The key to color image feature extraction is how to use the color information in the image more reasonably. Commonly used feature extraction methods include global methods, local methods, and methods based on deep learning. The global method directly describes the overall image and ignores the local information between adjacent pixels. This method is very sensitive to image changes such as illumination. ; The local method extracts features from the local area of the image, retains the locality of the data, bu...
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 IPC IPC(8): G06T7/90G06K9/46G06K9/62
CPCG06T7/90G06T2207/10024G06V10/56G06V10/462G06F18/22G06F18/241G06F18/214
Inventor 毕秀丽耿韬
Owner CHONGQING UNIV OF POSTS & TELECOMM