研究目的
To research the effectiveness of the main histogram-based methods for enhancing the integral contrast of complex monochrome images with low-contrast small-sized objects and non-uniform illumination in automatic mode, and to perform a comparative analysis of no-reference assessment methods for generalized contrast.
研究成果
Histogram-based methods can significantly increase the contrast of complex monochrome images with acceptable computational costs in automatic mode. Techniques based on histogram equalization (e.g., GHE, BBHE, DSIHE) provide the maximum increase in integral contrast but may reduce contrast for small objects. Methods like power-law intensification and adaptive contrast enhancement are effective and recommended for automatic enhancement without such drawbacks. Future work should focus on optimizing parameter selection and handling diverse image types.
研究不足
The effectiveness of methods like linear and non-linear stretching depends heavily on the initial image structure and parameter values, which may not be universally optimal. Histogram equalization techniques can suppress contrast for small-sized objects, reducing informativity. The study is limited to monochrome images and specific test cases, potentially not generalizing to all image types.
1：Experimental Design and Method Selection:
The study involved a comparative analysis of ten known histogram-based methods for contrast enhancement, including linear stretching, sigmoid function, non-linear stretching, gamma correction, global histogram equalization (GHE), brightness preserving bi-histogram equalization (BBHE), dualistic sub-image histogram equalization (DSIHE), power-law intensification, and a method based on brightness distribution parameters.
2：Sample Selection and Data Sources:
Four test monochrome images with complex structures, low-contrast small-sized objects, and non-uniform illumination were used.
3：List of Experimental Equipment and Materials:
No specific equipment or materials are mentioned; the methods are computational and applied to digital images.
4：Experimental Procedures and Operational Workflow:
Each test image was processed using the ten methods, and the results were evaluated using no-reference metrics for integral contrast (weighted and linear contrast definitions) and expert assessments.
5：Data Analysis Methods:
The effectiveness was measured using metrics such as complete and incomplete integral contrast based on weighted and linear definitions, and results were compared through graphical analysis and expert evaluations.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
