图像的修复方法和干细胞图像的分割方法
By constructing an adjustment parameter matrix to correct grayscale values and combining double threshold segmentation and hole filling methods, the problems of halo artifacts, low contrast, and uneven illumination in stem cell image segmentation were solved, and accurate stem cell image segmentation was achieved.
CN114648464BActive Publication Date: 2026-07-17SHANGHAI INSTITUTE OF TECHNICAL PHYSICS CHINESE ACADEMY OF SCIENCES +1
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI INSTITUTE OF TECHNICAL PHYSICS CHINESE ACADEMY OF SCIENCES
- Filing Date
- 2022-04-05
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Traditional image segmentation algorithms cannot accurately segment stem cell images, especially due to problems such as halo artifacts, low contrast, uneven lighting, and cell adhesion, which cause threshold segmentation methods to fail. In addition, neural network training datasets are required and computation is complex.
Method used
By segmenting the image into sub-blocks, constructing an adjustment parameter matrix to correct grayscale values, and combining double threshold segmentation and hole filling methods, stem cell image repair and segmentation can be achieved.
Benefits of technology
It improves the illumination uniformity of images, achieves accurate segmentation of stem cell images, and reduces computational complexity and dataset requirements.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN114648464B_ABST
Abstract
本发明提供了图像的修复方法和干细胞图像的分割方法，所述图像的修复方法包括以下步骤：(A1)将尺寸为M×N图像分割为尺寸为m×n的多个子块，M＝m×s，N＝n×t；(A2)获得所述图像的平均灰度值μ，以及第c行、第d列个子块的平均灰度值μcd，从而获得与第c行、第d列个子块对应的比值c＝1,2···m，d＝1,2···n；(A3)构建调整参数矩阵R′；(A4)对于矩阵R′中以rcd为中心的尺寸为(2l+1)×(2l+1)的邻域范围Ω，得到比值rcd的校正值r′cd；(A5)得到第c行、第d列个子块中像素点的灰度值p(i,j)校正值p′(i,j)＝p(i,j)×r′cd，从而得到修复后图像。本发明具有修复效果好等优点。
Need to check novelty before this filing date? Find Prior Art