中文摘要：

针对红外图像对比度低、边沿模糊、图像噪声明显以及去噪算法复杂度普遍高的缺点,提出应用于红外图像去噪的改进的引导滤波算法,并与传统的引导滤波去噪算法进行对比,以验证该算法提高图像质量的能力与算法处理的效率.该方法利用Sobel滤波代替传统引导滤波求统计特征的方法来获取图像特征,在保证去噪效果的前提下,极大简化了算法复杂度,并移植该算法至TMS320C6748DSP平台进行测试.实验结果表明,程序经优化后,达到实时处理的要求.

英文摘要：

Due to the low contrast,blurred edge and insufficient image noise of infrared images,and the highly complex denoising algorithms,the paper proposes an improved guided filter algorithm applied in infrared image denoising. Compared with the traditional guided filter denoising algorithm,the paper validates the ability and the efficiency of this algorithm to improve the quality of the image. The method adopts Sobel filter instead of the traditional guided filter for the statistical characteristics of the image to obtain the image features so as to greatly simplify the complexity of the algorithm while assuring the denoising effects. The algorithm is transplanted to the TMS320C6748 DSP platform for testing.After optimization,the algorithm meets the requirements of real-time processing.