针对煤矿井下无线传感网络因信息传输量大而导致传感节点能量消耗快、设备寿命缩减的问题,在分析煤矿井下图像特点和小波变换系数层特点的基础上,提出一种基于小波变换的压缩感知稀疏图像处理算法,即分块稀疏算法.采用bior4.4小波变换对煤矿井下图像信号进行分块稀疏变换,通过测量矩阵对稀疏化的图像数据进行测量,得到测量数据,再利用ROMP算法对稀疏变换的图像进行重构.仿真试验结果表明,与传统的压缩感知算法相比,该算法不仅能够以更低的采样率获得高质量的重构图像效果,压缩了图像大小,且能够快速地还原图像.%According to a quick energy consumption of the sensor nodes caused by the high information transmission,the reduced equipment service life and other problems of the wireless sensor network in the underground coal mine,based on the analysis on image features and wavelet transform coefficient layer features of the underground coal mine,a processing algorithm of the compressed perception and sparse images was provided based on the wavelet transform that was blocking sparse algorithm.A bior4.4 wavelet transform was applied to the blocking sparse transform on the images and signals in the underground coal mine.The measurement matrix was applied to the measurement of the spares image data,the measured data were obtained and then the ROMP algorithm was applied to the restructure of the spares transform images.The simulation experiment results showed that in comparison with the traditional compressed perception algorithm,the processing algorithm could not only obtain the high quality restructured image effect with a lower sampling rate,but could compress the image size and could rapidly restore the images.
展开▼
