资源列表
codefans.net
- 一个简单的车牌识别源码,可以识别蓝牌 白牌等国内车辆-A simple source license plate recognition
CamImag_baidu
- opencv244+vs2010摄像程序,基于MAT结构,带取景框,存盘功能-opencv244+vs2010 camera shooting program, base in MAT structure
opencv
- 关于运动目标的跟踪检测算法,CAMSHIFT经典算法,对研究这一领域的技术十分重要-Tracking moving targets detection algorithm, CAMSHIFT classical algorithm, the study of technology in this field is very important
Moving-target-detection--opencv
- 运动目标自动跟踪检测,选择一块区域,根据一些色差值等,检测相关点-Moving target tracking and detection automatically select an area, according to some color difference detection point
KinectBridgeWithOpenCVBasics-D2D
- 基于opencv的骨骼检测,可以很好地完成单人多人的检测-Based on the opencv bone detection, it is very effective for the detection of single people
ahe
- 自使用直方图均衡AHE的opencv实现,可以用于减弱光照影响和雾天去雾方面的研究,欢迎下载-Opencv realized from the use of histogram equalization AHE can be used to weaken the study of light effects and fog to fog, welcome to download
Erzhi
- OPENCV中关于图像的二值化处理的程序-The OPENCV on image binarization processing procedures
OpenSURF-H
- 关键点检测和描述算法,surf算法的opencv的源码,surf算法的性能与sift相当,但速度要远大于sift算法,经过测试可以正常运行-Keypoint detection and descr iption algorithm, algorithm source code at the Surf, surf the performance of the algorithm and sift quite, but the speed is much larger than sift algor
360_panorama
- 360-panorama,360度柱面图像拼接,程序在VS2008环境下开发,opencv2.4.3版本,程序可以直接运行,以将所需依赖项添加在文件夹中,源图像放在source_pic中,最后生成结果在result_pic中。 实现步骤:1先将图像投影到柱面坐标上,2提取图像SIFT描述子,3将SIFT描述子配对,4用RANSAC算法根据配对成功的描述子统计出变换矩阵T,5图像拼接。 对于拼接后的图像,没有做进一步细化处理,使用者可以在生成的图像上再做进一步编辑。-360-panoram
testthin
- 图像细化。将一幅图片进行细化,并输出结果-Image thinning. A picture refinement, and outputs the result
getMoravecCorners
- Moravec角点检测。算法会检测图像的每一个像素,将像素周边的一个邻域作为一个patch,并检测这个patch和周围其他patch的相关性-Moravec corner detection. Algorithm will detect each of the pixels of the image, as a neighborhood of pixels surrounding a patch, and detecting the correlation of this patch and p
searchTarget
- 目标匹配。关于匹配方法,使用不同的方法产生的结果的意义可能不太一样,有些返回的值越大表示匹配程度越好,而有些方法返回的值越小表示匹配程度越好-Target match. Matching method, the significance of the results generated by using different methods may not be the same, some returned larger value indicates a better match extent
