研究目的
To develop a novel game theory inspired binarization technique for degraded document images that outperforms state-of-the-art methods.
研究成果
The proposed GiB method achieves promising results on seven publicly available datasets, outperforming state-of-the-art methods in many cases. It effectively handles various types of document degradation through a combination of game theory and K-means clustering.
研究不足
The method's performance is dependent on the quality of the input images and may require parameter tuning for different types of document degradation.
1：Experimental Design and Method Selection:
A two-player, non-zero-sum, non-cooperative game is designed at the pixel level to extract local information, which is then fed to a K-means algorithm for classification.
2：Sample Selection and Data Sources:
Seven publicly available datasets (DIBCO 2009-14 and 2016) are used for testing.
3：List of Experimental Equipment and Materials:
Not explicitly mentioned.
4：Experimental Procedures and Operational Workflow:
Includes background estimation and image normalization, binarization using the game theory inspired technique, and post-processing to refine results.
5：Data Analysis Methods:
Performance is evaluated using precision, recall, F-measure, pseudo F-measure, peak signal-to-noise ratio, and distance reciprocal distortion metric.
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
