OpenCV 背景减法学习率不能改变 [英] OpenCV background subtraction learning rate cannot change

问题描述

我希望用 50 帧训练一个背景区域，并使用这个预先训练的模型进行背景减法.模型在训练后停止更新.

I wish to train a background region with 50 frames and use this pre-trained model for background subtraction. Model stops updating after training.

这是我的代码

import cv2
print "This program is for background subtraction with pre-trained model\n"

Training_Floder = "/Users/yuyang/Desktop/img1/"
Start_Frame_Num = 1
End_Frame_Num = 51

cv2.namedWindow("BG_IMAGE")

fgbg = cv2.createBackgroundSubtractorMOG2(50, 16, False)
font = cv2.FONT_HERSHEY_SIMPLEX



for index in range(Start_Frame_Num, End_Frame_Num):
    Img_File_Name = Training_Floder + str(index) + ".jpg"
    Img = cv2.imread(Img_File_Name)
    fgmask = fgbg.apply(Img, -1)
    BG_IMG = fgbg.getBackgroundImage()
    #######
    cv2.putText(BG_IMG,str(index),(10,500), font, 1,(255,255,255),2)
    cv2.imshow("BG_IMAGE", BG_IMG)
    #######
    cv2.waitKey(0)

Testing_Floder = "/Users/yuyang/Desktop/New/"
Test_Start = 1
Test_End = 100

for index in range(Test_Start, Test_End):
    Img_File_Name = Testing_Floder + str(index) + ".jpg"
    Img = cv2.imread(Img_File_Name)
    fgmask1 = fgbg.apply(Img, 0)
    BG_IMG1 = fgbg.getBackgroundImage()
    cv2.putText(BG_IMG1,str(index),(10,500), font, 1,(255,255,255),2)
    cv2.imshow("BG_IMAGE", BG_IMG1)
    cv2.waitKey(0)

根据评论

学习率参数在函数apply()"中.

The learning rate parameter is in the function "apply()".

@param learningRate 
The value between 0 and 1 that indicates how fast the background 
model is learnt. Negative parameter value makes the algorithm to 
use some automatically chosen learning rate. 0 means that the 
background model is not updated at all, 1 means that the background 
model is completely reinitialized from the last frame.

CV_WRAP virtual void apply(InputArray image, OutputArray fgmask, double learningRate=-1) = 0;"

但是，我在这里尝试了几种学习率:

However, I tried several learning rate here:

fgmask = fgbg.apply(Img, -1) or
fgmask = fgbg.apply(Img, 0) or
fgmask = fgbg.apply(Img, 1) or
fgmask = fgbg.apply(Img, 0.00001)

训练背景结果没有改变.这意味着我不能在测试时保持预训练模型不变！

The training Background result does not change. This means I CANNOT keep pre-trained model unchanged while testing!

我的代码有问题吗?有没有办法改变学习率?

Is there anything wrong with my code? Is there any way to change the learning rate?

这是一些结果

测试图像#1的背景减法结果

测试图像#40的背景减法结果

从上面的结果可以明显看出，虽然我将学习率设置为 0，但训练后的背景图像在测试时会发生变化.

From the result above, it is clear that the trained background image changes while testing, although I set learning rate as 0.

fgmask1 = fgbg.apply(Img, 0)

推荐答案

所以使用python实现的正确方法是

So the correct way to use the python implementation is

fgbg = cv2.createBackgroundSubtractorMOG2(50, 16, False)
fgbg.apply(input, output, learning_rate)

与在 C++ 实现中完全一样.学习率必须是第三个参数.

Exactly as in the c++ implementation. The learning rate must be the third parameter.

这篇关于OpenCV 背景减法学习率不能改变的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！