qiniudrive-master

说明：  在LS-SVM的基础上进行优化求解，速度更快更优。(On the basis of LS-SVM, the optimization solution is faster and better.)

读取显示BMP图像(read BMP images show)

何恺明等人研究出的基于暗通道的经典图像去雾算法，不仅可以还原图像的颜色和能见度，同时也能利用雾的浓度来估计物体的距离。(The classic fog removal algorithm based on dark channel, which was developed by He Kaiming and others, not only can restore the color and visibility of images, but also can estimate the distance of objects by using the concentration of fog.)

说明：  我们提出了一个简单、灵活和通用的对象实例分割框架。我们的方法能有效检测图像中的对象，同时为每个实例生成高质量的 segmentation mask。这种被称为 Mask R-CNN 的方法通过添加用于预测 object mask 的分支来扩展 Faster R-CNN，该分支与用于边界框识别的现有分支并行。Mask R-CNN 训练简单，只需在以 5fps 运行的 Faster R-CNN 之上增加一个较小的 overhead。此外，Mask R-CNN 很容易推广到其他任务，例如它可以允许同一个框架中进行姿态估计。我们在 COCO 系列挑战的三个轨道任务中均取得了最佳成果，包括实例分割、边界对象检测和人关键点检测。没有任何 tricks，Mask R-CNN 的表现优于所有现有的单一模型取得的成绩，包括 COCO 2016 挑战赛的冠军。(Mask R-CNN code by HeKaiming)

说明：  python语言，三种阈值分割方式，简单阈值分割，自适应阈值分割，OTsu分割方式(Python language, three methods of image segmentation , simple threshold segmentation, adaptive threshold segmentation, OTsu segmentation)

说明：  张正友标定方法的代码，已经包含了优化等情况，直接输入变可获得(calibration method code, Optimization has been included, direct input variable can be obtained)

说明：  FastLib图像特效库（新）.一些图形方面的算法如翻转，放大缩小，都有例子：）(FastLib Image effects for the (new). Some graphics algorithms such as overturned, zooming, Some examples :))

说明：  铁轨线检测，可以完成对直轨和弯轨的检测，效果较好(Rail line detection, can complete the detection of straight rail and curved rail, the effect is good)

实现通过用户给定B样条曲线的控制顶点、设定B样条曲线的次数及类型，进而绘制出相应的B样条曲线和对应的基函数图形。并且可以通过鼠标拖拽控制顶点修改B样条曲线和对应的基函数图形。从而，对B样条曲线和基函数的性质有更深入更直观的了解，理解B样条曲线和基函数的对应关系，在以后的学习中能更好地运用B样条曲线。(The number and type of user to achieve a given control vertices B-spline curve, set the B-spline curves, and then draw the corresponding B-spline basis functions and the corresponding graphic. And can be modified by dragging the mouse to control the apex B-spline curves and the corresponding basis function graphics. Thus, the nature of the B-spline curves and basis functions have a deeper and more intuitive understanding, understanding of the correspondence between the B-spline curves and basis functions in later learning to better use of B-spline curves.)

在linux系统中通过卷积神经网络实现图像风格迁移(Image style migration through convolution neural network in linux)