内容：
针对电气设备热故障定位采用彩色图像融合方法时数据量大的问题,对彩色图像融合以及压缩感知理论进行了研究和归纳,提出了一种基于压缩感知的红外与可见光彩色图像加权融合算法。首先分别将红外热图像与可见光彩色图像分解为单独的3个R、G、B通道,运用小波基对分离出来的各通道数据进行了稀疏表示;其次利用高斯随机矩阵对稀疏数据进行采样得到了测量值,将对应通道的测量值通过两个权重因子进行了融合;最后通过正交匹配追踪重构算法(OMP)对各通道的融合数据进行了重构,将得到的3通道重构数据恢复为融合后的彩色图像。研究结果表明,该算法能用比传统彩色图像融合少50%的采样点来实现融合;在平均梯度、峰值信噪比、空间频率以及信息熵上与传统数据融合指标相当,较好地保留了红外热图中的温升区域信息和可见光背景信息。Aiming at the problem of large amount of data in the thermal image fusion of electrical equipment, the color image fusion and compression perception theory are studied and summarized. A weighted weighted fusion algorithm of infrared and visible color image based on compression perception is proposed. Firstly, the infrared image and the visible color image are decomposed into three separate R, G and B channels respectively. The data of each channel separated by the wavelet base pair are sparse. Secondly, the sparse data is sampled by Gaussian random matrix. And the measured values of the corresponding channel are merged by two weighting factors. Finally, the fusion data of each channel are reconstructed by orthogonal matching tracking and reconstruction algorithm (OMP), and the resulting 3-channel reconstruction data is restored For the fusion of color images. The results show that the algorithm can be fused with less than 50% of the sampling points of traditional color images. The average gradient, peak signal to noise ratio, spatial frequency and information entropy are equivalent to those of traditional data fusion. Infrared thermal map of the temperature rise area information and visible light background information. 

文件列表（点击上边下载按钮，如果是垃圾文件请在下面评价差评或者投诉）: