Efficient adaptive multi-threshold image segmentation method

Based on the valley bottom characteristics of the image histogram, combined with dynamic programming and division and governance algorithm, the threshold search problem is transformed into matrix search, which solves the problem of threshold segmentation class specification and large search space in the existing technology, and realizes efficient adaptability image segmentation.

CN118967733AActive Publication Date: 2024-11-15CHONGQING UNIV OF TECH
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410985980.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-23
Publication Date
2024-11-15
Estimated Expiration
2044-07-23

AI Technical Summary

Technical Problem

The existing threshold segmentation method requires artificially specifying a fixed threshold segmentation category, and the threshold search space is large, resulting in high computational complexity and poor real-time performance.

Method used

Based on the valley characteristics of the image histogram and global optimization method, dynamic programming algorithm and division and conquer algorithm are used to convert threshold search problems into matrix search problems, and SMAWK algorithm reduces the complexity of matrix search time, and conducts global searches in the reduced search space. to determine the optimal number of thresholds and thresholds.

Benefits of technology

It effectively reduces the threshold search space, improves segmentation efficiency and accuracy, realizes adaptive image segmentation, and significantly reduces time complexity and running speed.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118967733A_ABST
    Figure CN118967733A_ABST
Patent Text Reader

Abstract

The invention provides an efficient self-adaptive multi-threshold image segmentation method, which comprises the following steps of: screening image gray levels by utilizing the characteristic that a segmentation threshold usually appears at a valley bottom position of an image histogram, and screening an alternative threshold from the image gray levels, thereby reducing a threshold search space and limiting the threshold search space within a valley range; based on a reduced threshold search space limited in a trough range, a threshold search problem solved by using a dynamic programming algorithm and a divide-and-conquer algorithm thought is converted into a matrix search problem, and the matrix search time complexity is reduced to a linear time complexity level in combination with an SMAWK algorithm; global search of different threshold numbers is carried out in the reduced threshold search space, the global optimal threshold number and the corresponding optimal threshold are determined, the situation that the threshold segmentation class number is appointed manually in advance is effectively avoided, and self-adaptive segmentation is achieved. According to the method, the threshold search space and the search complexity can be effectively reduced, and the feasibility and the applicability of threshold segmentation are improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Cited By

  • Breast cancer pathological image segmentation method

    CN119379728A