资源列表
green-screen
- 结合微软的Kinect传感器采集深度图像，将视野范围内的人物抠出来，与背景图叠加，相当于PS技术中的抠图。-Combined with Microsoft s Kinect sensor acquisition depth image, the characters within the field of view to pull out, and the background image overlay, equivalent to PS matting techniques.
object-measurement
- 利用微软的Kinect的深度数据，检测目标障碍物的大小尺寸，同时也可实现对人物身高的测量。-The use of Microsoft' s Kinect depth data, the detection target obstacle sizes, but also to realize the measurement of the height of the characters.
motion-detection
- 对进入摄像头范围内的物体进行捕捉，一旦有物体进入，则自动开启监控录像功能。-Within the object into the camera capture range, once an object enters, then automatically open video surveillance capabilities.
upload_5xsoft
- 图片上传的类-images process
DigitIdent
- 要求给定数字模式样本(7X9点阵，0-9共10个数字)，设计一BP神经网络进行手写体数字识别。 网络只有一个隐含层，训练样本100个，测试样本300个 -Requirements for a given sample in digital mode (7X9 lattice, a total of 10 figures 0-9), design a BP neural network to recognize handwritten digits. Network has only on
Snake
- 要求给定数字模式样本(7X9点阵，0-9共10个数字)，设计一BP神经网络进行手写体数字识别。 网络只有一个隐含层，训练样本100个，测试样本300个 -Requirements for a given sample in digital mode (7X9 lattice, a total of 10 figures 0-9), design a BP neural network to recognize handwritten digits. Network has only on
blog_code
- 针对10个数字的识别问题，本文利用BP 神经网络进行了实验-For 10-digit identification problem, we use BP neural network experiments
finspancurve
- 可以将图形图像，转换成灰度图像，根据灰度图像识别出图像内的高曲率点，以及图形的极值点，得出极值点曲率以及曲率半径-a graphic image can be converted into a grayscale image, a grayscale image according to the high curvature points identified within the image, and the extreme point pattern, derived extrema of
finspancurve
- 可以将图形图像，转换成灰度图像，根据灰度图像识别出图像内的高曲率点，以及图形的极值点，得出极值点曲率以及曲率半径-a graphic image can be converted into a grayscale image, a grayscale image according to the high curvature points identified within the image, and the extreme point pattern, derived extrema of
Unwrapping
- 在结构光测量了系统中，对光栅图形进行相位去包裹处理-zai jie gou guang ce liang xi tong zhong phase unwrapping
ACO
- 该文档包含代码和算法文档，主要用蚁群算法和互信息刚性配准多模图像-This document contains the code and algorithms documents, mainly with ant colony algorithm and mutual information-rigid registration multimodal image
resize
- 实现bmp格式图片的边界剪裁，即利用四个边界阈值来去掉边界区域，保留中心部分图像-this code is written to resize the bmp pictures