Edge detection method based on underwater image statistical property
A technology of underwater images and statistical characteristics, applied in the field of image processing, can solve problems such as incomplete structural information and false edges in noisy images.
Inactive Publication Date: 2015-02-25
三亚哈尔滨工程大学南海创新发展基地
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Sobel operator and Robert operator are faster in edge detection, but the detected structure information is incomplete, and false edges often appear in noise image detection
The wavelet method takes advantage of time-frequency analysis to obtain image edge information at different scales, but the wavelet method is limited to only detecting point singularity and limited directionality
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
[0043] The present invention is described in more detail below in conjunction with accompanying drawing example:
[0044] combine figure 1 ~7, the edge of the image is generally the place where the grayscale of the image changes drastically. There are many corresponding image edge detection methods, and the Canny algorithm is considered to be one of the most successful and widely used edge detection methods. Therefore, the present invention combines the Rosin threshold Select the method for Canny edge detection. It is mainly divided into the following stages:
[0045] (1) Analyze the statistical characteristics of underwater images
[0046] When analyzing the statistical properties of underwater images, it is assumed that the statistical properties of underwater images are ergodic. The present invention analyzes multiple images, and finds that the gray scale probability distribution of underwater images is unimodal. In this paper, two underwater images are selected, such as...
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 aims at providing an edge detection method based on an underwater image statistical property. The method includes the following steps: statistical property analysis is conducted on an underwater image, two underwater images are smoothened, a gradient amplitude and a gradient direction of a smoothened data array are calculated, non-maximum restrain is conducted on the gradient amplitude, an unimodal image threshold is obtained through a Rosin method, and edges are searched in a low-threshold image continuously by using a recursion tracking algorithm until all gaps in a high-threshold image are connected. A reasonable threshold of edge detecting is set automatically according to different underwater images, and the defect of a tradition method of being incorrect in manual setting of the threshold is overcome. Simultaneously, by confirming gauss smooth parameters reasonably and capturing image edge information in a neighborhood range, an edge outline of an image can be detected well while noise is removed.
Description
technical field [0001] The invention relates to an image detection method in the technical field of image processing. Background technique [0002] The image edge usually corresponds to the area where the image intensity or texture changes. It is the most basic feature to describe the target structure and an important clue for visual perception. It is widely used in linear feature extraction, target recognition and feature matching. The digital processing technology of underwater images has been paid more and more attention due to the important needs of the national economy such as ocean development and underwater target detection. [0003] At present, there are few researches on edge detection methods specifically for the statistical characteristics of underwater images. Commonly used methods are edge detection methods for ordinary optical images, such as Sobel, Robert, Canny, wavelet modulus maximum method and other detection methods. Sobel operator and Robert operator ar...
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): G06T7/00
Inventor 李一兵刘悦叶方林云张静李一晨李骜朱瑶
Owner 三亚哈尔滨工程大学南海创新发展基地