Large-Scale Image Segmentation Algorithm Based on Multi-Class Maximum Variance Algorithm
A maximum variance, image segmentation technology, applied in the field of image processing, can solve the problems of moderate extraction of key feature information of difficult images, complex segmentation algorithm, low image resolution, etc., achieve obvious segmentation effect, improve accuracy, and improve image segmentation accuracy Effect
Active Publication Date: 2018-09-18
常微分信息科技(苏州)有限公司
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0017] The maximum variance between classes can be extended to multiple classes. However, there are two key issues when seeking the maximum variance for multiple classes. One is how to determine the number of categories, and the other is the complexity of multi-class maximum variance.
Restricted by the above two key problems, the existing image segmentation algorithm that uses the maximum variance between classes to process images is difficult to properly extract the key feature information of the image. high
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
[0039] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.
[0040] for figure 1 The large-scale image shown is a grayscale image of an actual road image taken by a high-definition camera. The large-scale image segmentation algorithm based on the multi-category maximum variance algorithm of the present invention includes the following steps:
[0041] (1) get figure 1 Then the scanned image is converted into a grayscale image, and then the grayscale image is analyzed by the multi-class between-class maximum variance algorithm to obtain the initial contour of interest {A1,A2,A3,...,An}, where A1, A2, A3,..., An are defined as all contour points that constitute the initial contour of interest;
[0042] (2) According to the initial contour of interest {A1, A2, A3,..., An}, the region of interest in the gray image is obtained by enveloping, and according to the shape of the region of interest, select a regio...
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 invention discloses a large-scale image segmentation algorithm based on the maximum variance algorithm among multiple classes. The large-scale image segmentation algorithm based on the maximum variance The grayscale image is analyzed to obtain the grayscale image area of interest, and the grayscale image area of interest is divided into non-overlapping sub-blocks. The current grayscale image is segmented by the region growing method, and all pixels are considered comprehensively. The accuracy of pixel segmentation is improved, and the deficiency of the existing segmentation method is avoided; the segmentation algorithm of the invention can effectively segment large-scale images, the segmentation effect is obvious, and the image segmentation accuracy can be improved.
Description
technical field [0001] The invention belongs to the field of image processing, and in particular relates to a large-scale image segmentation algorithm based on a multi-category maximum variance algorithm. Background technique [0002] Image segmentation refers to the use of certain features in the image information to extract the user's interest from the image. Information analysis is performed on the image to be segmented, and feature extraction is performed on important information. Before the initialization segmentation, the features of each pixel in the image are first extracted. After feature extraction, a corresponding feature image can be obtained. This feature image contains three channels, and each channel corresponds to a set of feature values. However, for images of different scenes, their feature images have different characteristics. For example, for images with strong texture characteristics, their direction features will have greater contrast than the other t...
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): G06K9/32G06K9/34
CPCG06V10/25G06V10/267
Inventor 傅均汤旭翔陈柳柳曹海洋
Owner 常微分信息科技(苏州)有限公司