surf2stl

N-FINDR算法 端元提取 读取MNF降维后的影像数据 读取MNF降维后的影像数据( using N-FINDR end-member extraction)

用于图像去噪的代码，有模型去噪、双树复小波去噪，希望有帮助。(Code for image denoising, a denoising model dual tree complex wavelet denoising, hope to have help。)

NCUT切割，对源码进行了改进，改进后取消了只能压缩到160*160才能切割的限制(NCUT cutting, source code improvements, improved to cancel restrictions can only be compressed to 160* 160 to cut)

说明：  SDK_key: 人脸追踪(FT) Key 5qTrdeA43zdBXJZt7HM1x1wX9XGt7QicRU6Nt8rXLQAn 人脸检测(FD) Key 5qTrdeA43zdBXJZt7HM1x1weJvY2koYyzGARH6wVoTVw 人脸识别(FR) Key 5qTrdeA43zdBXJZt7HM1x1wmUKoEpvhgJYHd5xAsFig3 年龄识别(Age) Key 5qTrdeA43zdBXJZt7HM1x1xPHL78Ekcog6LU5iNCCNLj 性别识别(Gender) Key 5qTrdeA43zdBXJZt7HM1x1xWSjNHHcbTHHP9VuHLX1rf 此处APP_Id与SDK_Key仅适用于ArcFace v1.1， 不适用于其他版本。(DK_key: Face Tracking (FT) Key 5qTrdeA43zdBXJZt7HM1x1wX9XGt7QicRU6Nt8rXLQAn Face Detection (FD) Key 5qTrdeA43zdBXJZt7HM1x1weJvY2koYzGARH6wVoTVw Face Recognition (FR) Key 5qTrdeA43zdBXJZt7HM1x1wmUKoEpvhg JYHd5xAsFig3 Age Identification (Age) Key 5qTrde A43zdBXJZt7HM1x1xPHL78Ekcog 6LU5iNCCNLj Gender Key 5qTrdeA43zdBXJZt7HM1x1xWSjNHcbTHHP9VuHLX1rf Here APP_Id and SDK_Key are only applicable to ArcFace v1.1, not to other versions.)

单张图像二值化程序，将单张的彩色图像处理成二值化图像(Single image two value program)

一些常用的图像处理评价指标，处理后图像对比度，边缘剩余数，处理前后图像的相似性....(Some commonly used image processing evaluation, contrast, edges remaining number of similarities ....)

图像分割--用matlab实现图像的自动多阈值分割 (Image segmentation- Image by matlab automatic multi-threshold segmentation)

图像平移处理，可以图片进行平移变换处理，提供演示图片(Image translation process, it can image pan transformation process provides demonstration image)

线性同余伪随机数发生器、小数开方伪随机数发生器、利用Logistic方程和混合光学双稳模型产生混沌序列产生的序列矩阵(Linear congruence pseudo-random number generator, decimal square pseudo-random number generator, using Logistic equation and hybrid optical bistable model to generate sequence matrix generated by chaotic sequence)

DPM作者获得 2010年CVPR终生成就奖，DPM是在HOG的基础上进行改进，解决了HOG算子的不足，并且对SVM也进行了小幅度改善提出了LSVM这个概念(DPM is improved on the basis of HOG. It solves the shortcomings of HOG operator, and makes a small improvement on SVM. The concept of LSVM is proposed)