chaosmotor

基于压缩感知的图像恢复重建的实验源代码以及相关图片及文件(Compressed sensing image reconstruction experiment source code and related images and documents)

强大的小波包分析程序，将信号分频分析，在感兴趣的频率域内抽取信号信息。(Powerful wavelet packet analysis procedures, the signal of frequency analysis, frequency domain of interest extracted signal information.)

说明：  关于图像处理技术在冬小麦测试中的应用，用VC编的软件能对小麦进行二值化处理，中值滤波和灰度变换。(On image processing technology in the winter wheat test the application of the software made by VC on wheat binarization processing, median filtering and gray-scale transformation.)

分别利用 Roberts、Prewitt和 Sobel 算子对原图像进行边缘检测和锐化。(Respectively use Roberts, Prewitt and Sobel operator on edge detection and sharpening the original image.)

需安装Halcon

1．选择一幅图像，对该图像进行DCT块操作（将图像分成8×8小块，对每个小块进行DCT变换）。 2.利用屏蔽矩阵对DCT系数进行区域选择，然后进行逆变换（IDCT），合成原图像，计算合成图像的PSNR。 3.选择一小波函数，对一幅图像进行分解，显示各变换系数，观察其规律，并进行小波逆变换，合成原图像。 (1. Select an image, the image DCT block operations (the image is divided into 8 × 8 small pieces of each small piece for DCT transform). 2. The use of shielding on the DCT coefficient matrix of regional choices, and then inverse transform (IDCT), synthesis of the original image, calculating the synthetic image PSNR. 3. To select a wavelet function, decomposition of an image to show the transform coefficients, to observe its laws, and inverse wavelet transform, the original synthetic image.)

KMEANS代码段，用于图像分割，已配置完成(KMEANS,use to achieve image segmentation)

利用压缩感知实现波达方向估计，运用奇异值分解对接收信号进行降维，再利用L1范数进行估计(Using compressed sensing DOA estimation using singular value decomposition to reduce the dimensionality of the received signal, and then using the L1 norm estimate)

本实验要求编写一个计算压缩-解压缩图像的均方根误差、均方信噪比的程序。该程序是一个通用程序。编写程序产生图示的结果。使用上面的保真度准则程序计算可视信息的损失特性。(Prepared in this experiment, a calculation of compression- decompression of the root-mean-square error of image, mean square signal to noise ratio of the procedure. The procedure is a common procedure. Picture shows the preparation of the results of procedures. The use of the above criteria for the fidelity of calculated loss characteristics of visual information.)

说明：  为了达到较好的隐藏效果，本文采用游程编码的方法在二值图像中隐藏信息。秘密信息嵌入时修改二值图像的游程长度，如果秘密信息为是0，则修改该游程长度为偶数；如果秘密信息的取值与游程长度的奇偶性相匹配，则不改变游程长度。提取时根据游程长度奇偶性提取出秘密信息。在信息隐藏时，改变游程的奇偶性时通过对下一个游程的第一个像素值进行修改，因此当下一个游程长度为1时通过改变当前有成的值实现。(In order to achieve better hiding effect, we use run length encoding method to hide information in two value images. When the secret information is embedded, the length of the travel length of the two value image is modified. If the secret information is 0, the length of the travel length is even even. If the value of the secret information is matched with the parity of the length of the run, the length of the journey is not changed. The secret information is extracted according to the parity of run length. In information hiding, changing the parity of the journey is modified by the first pixel value of the next run, so the current length is 1 by changing the current value.)