pinpufenxi

将图片转化为BMP位图，然后通过逐行扫描识别图片中的数字。(A picture into a BMP bitmap, and then through the progressive scanned image in the figure.)

HOG特征提取源代码，这份代码是针对行人检测编写的，可以根据个人需要设置参数。(HOG Feature Extraction source code, this code is written for pedestrian detection, you can set the parameters according to individual needs.)

基于CDD模型的快速数字图像修复算法 matlab版 (CDD model based on fast digital image restoration algorithm matlab version)

萤火虫算法的源代码，函数文件，可直接运行，用来寻找图像阈值分割的最优值（三阈值）。(firefly algorithm function.)

奇异值分解波束形成，自适应波束形成仿真。由均匀线阵形式，确定阵列的导向矢量；由阵列导向矢量，对接收信号进行建模仿真；根据约束条件，实现自适应波束形成算法；考察算法性能与信噪比，采样率，观测时间等参数的关系。 (SVD beamforming, adaptive beamforming simulation. Form by the ULA to determine the array steering vector by the array steering vector, the received signal modeling and simulation under constraint conditions, adaptive beamforming algorithm examine algorithm performance and signal to noise ratio, sample rate, observation time, etc. parameters.)

采用半隐式方案实现变分水平集图像分割方法(Active contour without edge, image segmentation)

求一个图像的最大内接矩形面积。 步骤： 1.相机标定。首先根据物像关系式求出其中的参数。注意参数求出后要进行参数校验。 2.从背景中分离出目标 3.边缘检测 4.目标形状参数计算。(For an image of the largest inscribed rectangle area. Steps: 1. Camera calibration. First of all object-image relationship in accordance with the parameters obtained. Attention to the parameters obtained after parameter calibration. 2. From the context of isolated target 3. Edge detection 4. Target shape parameters.)

说明：  对不同噪声强度的加噪图像，用粒子群优化算法优化结构元素，找到最优结构元素，对图像进行去噪操作，使恢复的图像达到最大峰值信噪比(Noise intensity for different images with noise, using particle swarm optimization algorithm to optimize the structure elements, find the optimal structure element, the image de-noising operation to restore the image to a peak signal to noise ratio)

基于颜色特征向量的粒子滤波跟踪程序，修复了因粒子超过窗口范围出现ERROR的BUG，还包括了一些有助于粒子滤波学习的资料！！(color-based particle filter tracking, some bugs have been fixed, some information about PF Tracking is included too!!)

利用OpenCV实现单幅图片对于手指的精确识别(Single image using OpenCV to achieve precise identification of the finger)