资源列表
ExThreshold
- 1、读取显示所给PET图像（SE2\IM210）。 2、学习imhist函数，画出所给PET图像的直方图。 3、根据直方图为该PET图像选择合适的显示窗（imshow参数），使胸 部轮廓等解剖结构可见。 4、根据直方图选择合适阈值，分割肿瘤区域与正常区域，计算肿瘤区 域内图像灰度均值、方差、最大值及最小值。-1 give PET images displayed, read (SE2 \ IM210). 2, learning, imhist function, draw
tuxiangpipei
- 图像配准利用各种方法面对各种问题来解决问题发现问题-Image registration using a variety of methods to solve problems face various problems problems found
canny-bianyuanjiance
- 改进的边缘检测算法，使用迭代法和腐蚀算法处理。-Improved edge detection algorithm, using an iterative algorithm processing method and corrosion.
qiuquzhixin
- 求取质心的两种代码，其中第二种包含求面积的一种方法，运行的结果一样。-Strike two codes centroid, wherein the second area comprises a method request, the results of running the same.
mubiaogenzong
- 本代码包含了多个动态目标的跟踪监测，其中背景差分法可以实现实时更新。-The code includes tracking and monitoring multiple dynamic objects, including background subtraction can be realized to date.
tiqutuxiangzhen
- 文件包含对视频提取每帧图像和保存图像到磁盘的代码。-Files contain extracts each frame of the video image and save the image to disk code.
diaoyongshexiangtou
- 文件包含调用摄像头并进行视频保存，可自行调整视频帧数和录制时间。-Files contain video call camera and saved, you can adjust their own video frames and recording time.
beijingchafenfa
- 背景差分法的两个例子，直接采用灰度图像，如是RGB图像请先自行灰度处理-Two examples of background subtraction, the direct use of gray-scale images, the case of RGB images, please self-gradation processing
hardexample
- 是指利用第一次训练的分类器在负样本原图上进行行人检测时所有检测到的矩形框，这些矩形框区域很明显都是误报，把这些误报的矩形框保存为图片，加入到初始的负样本集合中，重新进行SVM的训练，可显著减少误报。这种方法叫做自举法-The first is the use of trained classifier all detected rectangle detect pedestrians in the negative when the sample picture, the rectangular
picture-clipping
- 将图片随机剪裁成固定大小的图片，适用于机器学习中剪裁负样本-The image randomly cut to a fixed size image, for machine learning tailoring negative samples
Pyramid-lucas-kanade-optical-flow-
- 金字塔lucas-kanade光流代码，可用于动态背景，但速度较慢-Pyramid lucas-kanade optical flow codes, can be used for dynamic background, but slower
2D-linear-fitting
- 2D直线拟合，忽略了离群点，在测量值被高噪声，传感器故障等破坏的实际应用中十分重要-2D linear fitting, ignoring the outliers, are very important in high-noise measurements, sensor failures and other damage in the practical application