WAV-LSB-Watermark

数字图像处理中的基本形态学处理，如腐蚀、膨胀、开操作、闭操作等实现的源代码。可以自定义结构元素的形状。(Digital image processing in the basic morphology processing, such as corrosion, swelling, open operation and close operation to achieve the source code. You can customize the shape of structural elements.)

利用直方图双峰法选取适当阈值对图像进行二值化处理。此法只适用于直方图为双峰状图像。(Using the Shuangfeng method to the appropriate threshold histogram of image processing two values. This method is only applicable to the histogram of image of Shuangfeng.)

说明：  结合雷达RCS回波数据，在matlab中生成ISAR二维图像，以及距离像方位像(Combined with radar RCS echo data, ISAR two-dimensional image and range profile azimuth image are generated in MATLAB)

说明：  粒子滤波器matlab实现，用于实现目标检测算法(Particle filter matlab realized, for the achievement of Target Detection Algorithm)

说明：  fast与sift算法结合，实现特征匹配，可以参考；fast与sift算法结合，实现特征匹配，可以参考(The combination of fast and sift algorithm can achieve feature matching, which can be referred to; the combination of fast and sift algorithm can achieve feature matching, which can be referred to)

图像修复的一些常用模型，与tv模型，cdd模型，bscb模型，还有基于纹理的修复方法。(Some common model image restoration, and tv model, cdd model, bscb model, as well as repair method based on texture.)

针对红外图像特别是红外弱小目标的特点，提出一种基于双平台的直方图均衡算法。(The characteristics of the infared dim-small targets iamge was analyzed.A new contrast enhancement algorithm was presented based on double plateaus histogram equalization)

图像检测的算法，Faster R-CNN算法，先对整张图像进行卷积计算，然后通过感兴趣区域池化层(RoI Pooling Layer)将选择性搜索算法推荐出来的候选区域和卷积网络计算出的特征映射图进行融合，得到候选区域对应的特征矢量，这种共享卷积计算的操作极大地减少了卷积计算的次数。而且这些特征矢量的维度统一，方便后续的分类工作。通过感兴趣区域池化层处理卷积特征，并将得到的特征送往两个并行计算任务进行训练，分类和定位回归。通过这些方法和改进的框架，Fast R-CNN 用更短的训练和测试时长，取得了比 R-CNN 更好的效果(Faster R-CNN algorithm first convolutes the whole image, then fuses the candidate regions recommended by the selective search algorithm and the feature mapping maps calculated by the convolution network through the RoI Pooling Layer to get the corresponding feature vectors of the candidate regions, which greatly reduces the number of convolution calculations. Moreover, the dimension of these feature vectors is unified, which facilitates the subsequent classification work. The convolution feature is processed by the pooling layer of the region of interest, and the obtained feature is sent to two parallel computing tasks for training, classification and positioning regression. Through these methods and improved framework, Fast R-CNN uses shorter training and testing time and achieves better results than R-CNN.)

说明：  给出了一种基于小波分形的图像有损压缩方法.首先，利用小波变换将图像分解为不同频率的分量，然后采用分形迭代函数系统、Huffman树等方法对它们进行编码.实验表明，该算法在保证还原图像质量的情况下，有较高的压缩比.(Image Compression Bassed on Wavelet and Fractal)

在桌面上显示图像，C++Builder精选编程学习源码，很好的学习资料。(Image display on the desktop, C++Builder select learning programming source code, very good learning materials.)