如何使用C库刷新Python程序中的内存?

时间:2012-05-09 11:48:26

标签: python windows opencv memory-leaks c-libraries

Python程序功能是通过网络摄像头找到最亮的灯光,并将X,Y坐标发送到arduino程序。这是用相机跟随太阳。

问题是内存会增加,直到程序冻结。 我试图在使用后删除所有变量。但它仍然使用太多内存。

os:Windows 7

如何找到泄漏并停止泄漏?

这是python代码:

import cv
import traceback
import serial



try:
    cv.NamedWindow("CameraFeed", 1)
    capture = cv.CaptureFromCAM(0)
    cv.NamedWindow("Altered", 1)


except:
    cv.DestroyWindow("CameraFeed")
    cv.DestroyWindow("Altered")
    traceback.print_exc()

ser = serial.Serial("COM15",9600)



def repeat():
    imga = cv.QueryFrame(capture)
    img = cv.CreateImage(cv.GetSize(imga),8,3)
    cv.Smooth(imga, img, cv.CV_BLUR, 3)     


    thresholded_img =  cv.CreateImage(cv.GetSize(img), 8, 1)
    del(img)


    minv = cv.Scalar(250,250,250, 0)
    maxv = cv.Scalar(255,255,255, 0)
    cv.InRangeS(imga, minv, maxv, thresholded_img) 
    del(minv)
    del(maxv)        
            #determine the objects  moments and check that the area is large  
            #enough to be our object
    moments = cv.Moments(cv.GetMat(thresholded_img,1), 0) 
    area = cv.GetCentralMoment(moments, 0, 0)

            #there can be noise in the video so ignore objects with small areas 


    if(area > 10000):

                #determine the x and y coordinates of the center of the object 
                #we are tracking by dividing the 1, 0 and 0, 1 moments by the area 
            x = cv.GetSpatialMoment(moments, 1, 0)/area 
            y = cv.GetSpatialMoment(moments, 0, 1)/area 
            del(moments)
            del(area)
                #print 'x: ' + str(x) + ' y: ' + str(y) + ' area: ' + str(area) 

                #create an overlay to mark the center of the tracked object 
            overlay = cv.CreateImage(cv.GetSize(imga), 8, 3) 

            cv.Circle(imga, (int(x), int(y)), 5, (255, 0, 0), 20) 
            cv.Add(imga, overlay, imga)
            del(overlay)
                #add the thresholded image back to the img so we can see what was  
                #left after it was applied 
            cv.Merge(thresholded_img, None, None, None, imga)

            print 'X: ',x
            print 'Y: ',y
            print ' '
            if x < 300:
                print "left"    
            if x > 340:
                print "Right"
            if y < 210:
                print "Up"
            if y > 250:
                print "Down"

            ser.write('%03i%03i\r\n'% (int(x),int(y))) 







    else:
            ser.write('%03i%03i\r\n'% (320,240)) 



    cv.ShowImage("CameraFeed", thresholded_img)
    cv.ShowImage("Altered", imga)
    del(imga)
    del(thresholded_img)


try:
    while True:
        repeat()
        if cv.WaitKey(10) == 13:
            break
except:
    cv.DestroyWindow("CameraFeed")
    cv.DestroyWindow("Altered")
    ser.close()
    traceback.print_exc()

cv.DestroyWindow("CameraFeed")
cv.DestroyWindow("Altered")
ser.close()
del(serial)
exit()

2 个答案:

答案 0 :(得分:3)

这是GetMat的一个已知错误 - &gt; http://code.opencv.org/issues/1579

答案 1 :(得分:1)

代码看起来正确(example how to do it)。没有必要删除变量;当repeat()返回时,它们都将被删除。

我的猜测是您正在使用的OpenCV版本中存在错误。尝试不同的版本。如果行为发生在最新版本中而不是旧版本,请考虑报告错误。