以提升串行视觉通信下的舰船图像质量为目标，研究串行视觉通信下舰船失真图像颜色校正方法。采用多线程通信技术，通过写线程、读线程以及监测线程的多线程合作，完成舰船图像的串行视觉通信。依据自适应颜色阈值，划分待传输的舰船失真图像为多个层次，针对各层次的舰船图像设置不同尺度的细节函数，依据细节函数提取舰船失真图像的颜色特征信息。依据所提取的颜色特征，划分舰船失真图像的亮区域与暗区域。利用统计特征匹配方法以及直方图匹配方法，分别完成亮区域与暗区域的颜色校正，合成完成颜色校正的暗区域与亮区域，获取舰船失真图像颜色校正结果。实验结果表明，该方法有效校正串行视觉通信下的舰船失真图像颜色，舰船失真图像的色度、饱和度、亮度改善效果明显。
In order to improve the ship image quality under serial visual communication, the color correction method of ship distortion image under serial visual communication is studied. Using multi-thread communication technology, the serial visual communication of ship image is completed by multi-thread cooperation of writer thread, reader thread and monitor thread. According to the adaptive color threshold, the ship distortion image to be transmitted is divided into multiple layers. The detail functions of different scales are set for each layer of ship image, and the color feature information of ship distortion image is extracted according to the detail function. According to the extracted color features, the light region and dark region of ship distortion image are divided. The statistical feature matching method and histogram matching method were used to complete the color correction of the bright and dark regions respectively, and the dark and bright regions were synthesized to obtain the color correction results of ship distortion images. Experimental results show that the proposed method can effectively correct the color of ship distortion image under serial visual communication, and the chroma, saturation and brightness of ship distortion image are improved obviously.
2023,45(2): 167-170 收稿日期：2022-10-18
DOI：10.3404/j.issn.1672-7649.2023.02.030
分类号：TP391
基金项目：2021年国家社会科学基金面上项目（21BTJ057）；2021年教育部产学合作协同育人项目（202102197002）
作者简介：周燕(1980-),女,硕士,讲师,研究方向为数理统计及数据挖掘
