centroid

说明：  Create hex map settigns

卷积神经网络（Convolutional Neural Network）是一个专门针对图像识别问题设计的神经网络。它模仿人类识别图像的多层过程：瞳孔摄入像素；大脑皮层某些细胞初步处理，发现形状边缘、方向；抽象判定形状（如圆形、方形）；进一步抽象判定（如判断物体是气球）。(Convolutional neural network is a neural network specially designed for image recognition. It imitates the multi-layer process of human recognition image: pupil ingestion of pixels; some cells in the cerebral cortex preliminary processing to find shape edges and directions; Abstract judgment of shape (such as circles and squares); and further Abstract judgment (such as judging objects are balloons).)

CZT变换进行信号频谱分析，解决了FFT法频域分辨不高的问题，源码开发，特别适用(Chirp Z Transform)

HOG LBP 在python matlab C++环境下的实现(Implementation of HOG LBP in Python matlab C++ environment)

初步的代码，只保存边界的声波逆时偏移 RTM

对灰度图像或RGB图像，进行基于边缘检测和区域生长的图像分割技术(On grayscale images or RGB images, for edge detection and region growing-based Segmentation)

detection different object in video an moves with high accuracy

说明：  通过利用“Otsu”方法对图像进行全局阈值分割处理。然后，将灰度图转化成二值图，为获得大豆颗粒计数和大小做准备。调用函数获得图像连通图的数量，并测量标注矩阵中每个标注区域的属性，获得每个标注区域的长度。最后通过寻找每个区域的重心数，即可统计出大豆的个数(The Otsu method is used to segment the image. Then, the gray-scale image is transformed into a binary image to prepare for the soybean particle count and size. Call the function to get the number of connected graphs, measure the attributes of each annotation region in the annotation matrix, and get the length of each annotation region. Finally, the number of soybean can be calculated by finding the number of barycenters in each region)

MATLAB 计算医学图像的配准，融合 基于SIFT方法的配准()

说明：  ROC曲线函数,横轴为虚警概率，纵轴为检测概率(ROC curve function, the horizontal axis is false alarm probability, and the vertical axis is detection probability)