RETINEX

PCNN源代码,此为简化PCNN模型的实现原代码。(the source code of pcnn)

高光谱遥感图像光谱解混的独立成分分析技术(Hyperspectral remote sensing image spectra unmixing independent component analysis)

说明：  基于sift特征提取的图像配准的几个案例(Several cases of image registration based on SIFT feature extraction)

在图像中加入椒盐噪声的源代码，编译后的可执行文件须带参数运行(the image Salt and Pepper to the noise source, the compiled executable file parameters required to bring Operation)

说明：  用于相位解缠，提供好的解缠方法和解缠效果，能够更好的得到真实值(For phase unwrapping, unwrapping methods provide good wrapped around the effect of reconciliation that can better be the true value)

进一步讨论怎么样进行虹膜眼睛的位置确认，区域生长法确定内圆轮廓(To further discuss how to confirm the position of the iris eye and determine the inner circle contour by region growth method)

一种利用图像中直线特征的方法来标定摄像机畸变参数的方法。该方法提取图像中多组不同方向的平行直线，并对其利用非线性最小二乘方法进行优化以得到最优的畸变参数，在优化过程中，采用多步渐进的策略。该方法不需要已知三维世界坐标，易于实现。实验表明，该方法有较高的准确率。(A camera lens distortion method is proposed, which uses the straight line features. Many parallel lines of different direction were extracted, and then were used to optimize the distortion parameters by nonlinear least square. The thought of step by step was added when the optimization method working. 3D world coordination is not need to know, and the method is easy to implement. The experiment result shows its high accuracy.)

红外焦平面探测器输出图像信噪比计算和分析(Infrared focal plane image detector output signal to noise ratio calculation and analysis)

Sobel算子就是一种基于方向导数掩模求卷积的方法。通过边缘检测算子可以检测到每个像素的领域并对灰度变化率进行量化，也包括方向的确定。该代码是基于sobel算法的检测程序。(Sobel operator is a directional derivative mask and convolution methods. To each pixel and gray level in the field rate of change can be detected through the edge detection operator to quantify, including direction. The code is based on sobel algorithm testing procedures.)

说明：  不管对于图像的编码还是解码，三次样条算法的性能都是很好的，然后由于计算量比较大使他的效率有所下降，在次提出了一个考虑算法质量的前提下的快速算法。(regardless of the coding for images or decoding, cubic spline algorithm performance are good, and then because of the comparison calculation of the efficiency Ambassador he declined, in a meeting of the algorithm to consider the quality of the fast algorithm.)