HeKaiming_hazeremoval_CN_Matlab

从眼动仪提取数据，根据高斯核函数得到感兴趣区域(Extract data from the eye tracker, based on the Gaussian kernel function to get the area of ​ ​ interest)

用于计算两幅图片之间的峰值信噪比，从而衡量图片重建质量(It used to calculate the peak signal to noise ratio between the two pictures, which measure the quality of the reconstructed image)

数据库是由10个人的7种正面表情组成的213幅灰度图像，图像是以大小为256256的8位灰度级存储的，格式为.tiff型，平均每个人每种表情有2到4张。( the JAFFE database[11] which is composed of 213 images of female facial expression corresponding to 10 distinct subjects. Each image is stored at a resolution of 256×256 pixels and 8-bit gray level. Each subject in the database is represented with 7 categories of expression (angry, disgust, fear, neutral, sadness, happiness and surprise). )

Aptina ASX340图像传感器设计线路图及datasheet(Aptina ASX340 image sensor design schematics and datasheet)

实现肺实质初始分割及采用滚球法（形态学开运算）进行修补肺实质轮廓凹陷。(Achieve initial segmentation and the use of the pulmonary parenchyma ball method (morphological opening operation) to repair the lung parenchyma contour depression.)

说明：  编程实现对bmp图像最大值，最小值，标准差，熵，各灰度值频率的计算(Programming to calculate the maximum value, minimum value, standard deviation, entropy and gray value frequency of BMP image)

基于LDA的人脸识别技术，从ORL数据库读数据，然后从中取一部分进行训练，一部分进行测试。(This code is LDA base face recognition programme. It reads nots faces from ORL database and the rest (noc-nots) are used as test. LDA_Performance shows the recognition performance. )

图像处理中ransac算法使用工具，同图复制篡改检测(The image processing algorithm ransac tool tamper detection with FIG copy)

采用大津阈值法检测车道线，效果还可以，有需要的可以看看(Using the Dajin threshold method to detect the lane line, the effect is still good, you can see if you need it.)

图片去噪：对一幅图像加入不同的噪声（随机点噪声、椒盐噪声等），选取不同的方法去噪，比如说邻域平均、中值滤波、图像迭加等，比较对于不同的噪声，不同的方法哪种更好(Image denoising: for an image by adding different noise (random-dot noise, salt and pepper noise, etc.), select a different method of denoising, for example, the neighborhood average, median filter, image superposition and so on, compared for different noise, different methods which better)