python类absdiff()的实例源码

def get_match_confidence(img1, img2, mask=None):
    if img1.shape != img2.shape:
        return False
    ## first try, using absdiff
    # diff = cv2.absdiff(img1, img2)
    # h, w, d = diff.shape
    # total = h*w*d
    # num = (diff<20).sum()
    # print 'is_match', total, num
    # return num > total*0.90
    if mask is not None:
        img1 = img1.copy()
        img1[mask!=0] = 0
        img2 = img2.copy()
        img2[mask!=0] = 0
    ## using match
    match = cv2.matchTemplate(img1, img2, cv2.TM_CCOEFF_NORMED)
    _, confidence, _, _ = cv2.minMaxLoc(match)
    # print confidence
    return confidence

def diff_rect(img1, img2, pos=None):
    """find counters include pos in differences between img1 & img2 (cv2 images)"""
    diff = cv2.absdiff(img1, img2)
    diff = cv2.GaussianBlur(diff, (3, 3), 0)
    edges = cv2.Canny(diff, 100, 200)
    _, thresh = cv2.threshold(edges, 0, 255, cv2.THRESH_BINARY)
    contours, _ = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_NONE)
    if not contours:
        return None
    contours.sort(key=lambda c: len(c))
    # no pos provide, just return the largest different area rect
    if pos is None:
        cnt = contours[-1]
        x0, y0, w, h = cv2.boundingRect(cnt)
        x1, y1 = x0+w, y0+h
        return (x0, y0, x1, y1)
    # else the rect should contain the pos
    x, y = pos
    for i in range(len(contours)):
        cnt = contours[-1-i]
        x0, y0, w, h = cv2.boundingRect(cnt)
        x1, y1 = x0+w, y0+h
        if x0 <= x <= x1 and y0 <= y <= y1:
            return (x0, y0, x1, y1)

def get_match_confidence(img1, img2, mask=None):
    if img1.shape != img2.shape:
        return False
    ## first try, using absdiff
    # diff = cv2.absdiff(img1, img2)
    # h, w, d = diff.shape
    # total = h*w*d
    # num = (diff<20).sum()
    # print 'is_match', total, num
    # return num > total*0.90
    if mask is not None:
        img1 = img1.copy()
        img1[mask!=0] = 0
        img2 = img2.copy()
        img2[mask!=0] = 0
    ## using match
    match = cv2.matchTemplate(img1, img2, cv2.TM_CCOEFF_NORMED)
    _, confidence, _, _ = cv2.minMaxLoc(match)
    # print confidence
    return confidence

def diff_rect(img1, img2, pos=None):
    """find counters include pos in differences between img1 & img2 (cv2 images)"""
    diff = cv2.absdiff(img1, img2)
    diff = cv2.GaussianBlur(diff, (3, 3), 0)
    edges = cv2.Canny(diff, 100, 200)
    _, thresh = cv2.threshold(edges, 0, 255, cv2.THRESH_BINARY)
    contours, _ = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_NONE)
    if not contours:
        return None
    contours.sort(key=lambda c: len(c))
    # no pos provide, just return the largest different area rect
    if pos is None:
        cnt = contours[-1]
        x0, y0, w, h = cv2.boundingRect(cnt)
        x1, y1 = x0+w, y0+h
        return (x0, y0, x1, y1)
    # else the rect should contain the pos
    x, y = pos
    for i in range(len(contours)):
        cnt = contours[-1-i]
        x0, y0, w, h = cv2.boundingRect(cnt)
        x1, y1 = x0+w, y0+h
        if x0 <= x <= x1 and y0 <= y <= y1:
            return (x0, y0, x1, y1)

def features(image, channel, levels=9, start_size=(1983, 1088), ):
    """
        Extracts features by down-scaling the image levels times,
        transforms the image by applying the function channel to 
        each scaled version and computing the difference between 
        the scaled, transformed versions.
            image :         the image
            channel :       a function which transforms the image into
                            another image of the same size
            levels :        number of scaling levels
            start_size :    tuple.  The size of the biggest image in 
                            the scaling pyramid.  The image is first
                            scaled to that size and then scaled by half
                            levels times.  Therefore, both entries in
                            start_size must be divisible by 2^levels.
    """
    image = channel(image)
    if image.shape != start_size:
        image = cv2.resize(image, dsize=start_size)

    scales = [image]
    for l in xrange(levels - 1):
        logger.debug("scaling at level %d", l)
        scales.append(cv2.pyrDown(scales[-1]))

    features = []
    for i in xrange(1, levels - 5):
        big = scales[i]
        for j in (3,4):
            logger.debug("computing features for levels %d and %d", i, i + j)
            small = scales[i + j]
            srcsize = small.shape[1],small.shape[0]
            dstsize = big.shape[1],big.shape[0]
            logger.debug("Shape source: %s, Shape target :%s", srcsize, dstsize)
            scaled = cv2.resize(src=small, dsize=dstsize)
            features.append(((i+1,j+1),cv2.absdiff(big, scaled)))

    return features

def composite(img1, img2, mask0):
    if mask0.shape[2] == 3:
        mask2 = cv2.cvtColor(mask0, cv2.COLOR_BGR2GRAY)
    else:
        mask2 = mask0[:]
    mask1 = np.ones((img1.shape[0], img1.shape[1], 3), np.uint8)
    mask1[..., 0] = mask2
    mask1[..., 1] = mask2
    mask1[..., 2] = mask2
    white = np.ones((img1.shape[0], img1.shape[1], 3), np.uint8)
    white[:] = (0, 0, 0)
    invmask = np.zeros((img1.shape[0], img1.shape[1], 3), np.uint8)
    invmask = cv2.absdiff(white, mask1)
    invmask = cv2.bitwise_not(invmask)
    output = np.zeros((img1.shape[0], img1.shape[1], 3), np.uint8)
    cv2.subtract(img2, invmask, dst=output)
    return output

def test_similar():
    from itertools import combinations
    from collections import defaultdict
    from heapq import heappush


    def sim1(img1, img2):
        h, w, d = img1.shape
        total = h*w*d
        diff = cv2.absdiff(img1, img2)
        num = (diff<10).sum()
        return num*1.0/total

    names = [os.path.join('scene', c) for c in os.listdir('scene')]
    imgs = dict(zip(names, map(cv2.imread, names)))
    diffs = defaultdict(list)

    for name1, name2 in combinations(names, 2):
        img1, img2 = imgs[name1], imgs[name2]
        similarity = sim1(img1, img2)
        # print 'diff', name1, name2, 'result is:', similarity
        heappush(diffs[name1], (-similarity, name2))
        heappush(diffs[name2], (-similarity, name1))

    for k, v in diffs.iteritems():
        print k, v[0][1], -v[0][0]

def get_mask(img1, img2, thresh=20):
    if img1.shape != img2.shape:
        return
    diff = cv2.absdiff(img1, img2)
    diff = np.mean(diff, axis=2)
    diff[diff<=thresh] = 0
    diff[diff>thresh] = 255
    mask = np.dstack([diff]*3)
    return mask

def test_similar():
    from itertools import combinations
    from collections import defaultdict
    from heapq import heappush


    def sim1(img1, img2):
        h, w, d = img1.shape
        total = h*w*d
        diff = cv2.absdiff(img1, img2)
        num = (diff<10).sum()
        return num*1.0/total

    names = [os.path.join('scene', c) for c in os.listdir('scene')]
    imgs = dict(zip(names, map(cv2.imread, names)))
    diffs = defaultdict(list)

    for name1, name2 in combinations(names, 2):
        img1, img2 = imgs[name1], imgs[name2]
        similarity = sim1(img1, img2)
        # print 'diff', name1, name2, 'result is:', similarity
        heappush(diffs[name1], (-similarity, name2))
        heappush(diffs[name2], (-similarity, name1))

    for k, v in diffs.iteritems():
        print k, v[0][1], -v[0][0]

def get_mask(img1, img2, thresh=20):
    if img1.shape != img2.shape:
        return
    diff = cv2.absdiff(img1, img2)
    diff = np.mean(diff, axis=2)
    diff[diff<=thresh] = 0
    diff[diff>thresh] = 255
    mask = np.dstack([diff]*3)
    return mask

def convert_to_linedrawing(self, luminous_image_data):
        neiborhood24 = numpy.array([[1, 1, 1, 1, 1],
                                    [1, 1, 1, 1, 1],
                                    [1, 1, 1, 1, 1],
                                    [1, 1, 1, 1, 1],
                                    [1, 1, 1, 1, 1]],
                                   numpy.uint8)
        dilated = cv2.dilate(luminous_image_data, neiborhood24, iterations=1)
        diff = cv2.absdiff(dilated, luminous_image_data)
        linedrawing = cv2.bitwise_not(diff)
        return linedrawing

def rgConspicuity(image):
    """
        Creates the conspicuity map for the sub channel `red-green conspicuity'.
        of the color channel.
    """
    def rg(image):
        r,g,_,__ = cv2.split(image)
        return cv2.absdiff(r,g)
    fs = features(image = image, channel = rg)
    return sumNormalizedFeatures(fs)

def byConspicuity(image):
    """
        Creates the conspicuity map for the sub channel `blue-yellow conspicuity'.
        of the color channel.
    """
    def by(image):
        _,__,b,y = cv2.split(image)
        return cv2.absdiff(b,y)
    fs = features(image = image, channel = by)
    return sumNormalizedFeatures(fs)

#def sumNormalizedFeatures(features, levels=9, startSize=(640,480)):

def makeNormalizedColorChannels(image, thresholdRatio=10.):
    """
        Creates a version of the (3-channel color) input image in which each of
        the (4) channels is normalized.  Implements color opponencies as per 
        Itti et al. (1998).
        Arguments:
            image           : input image (3 color channels)
            thresholdRatio  : the threshold below which to set all color values
                                to zero.
        Returns:
            an output image with four normalized color channels for red, green,
            blue and yellow.
    """
    intens = intensity(image)
    threshold = intens.max() / thresholdRatio
    logger.debug("Threshold: %d", threshold)
    r,g,b = cv2.split(image)
    cv2.threshold(src=r, dst=r, thresh=threshold, maxval=0.0, type=cv2.THRESH_TOZERO)
    cv2.threshold(src=g, dst=g, thresh=threshold, maxval=0.0, type=cv2.THRESH_TOZERO)
    cv2.threshold(src=b, dst=b, thresh=threshold, maxval=0.0, type=cv2.THRESH_TOZERO)
    R = r - (g + b) / 2
    G = g - (r + b) / 2
    B = b - (g + r) / 2
    Y = (r + g) / 2 - cv2.absdiff(r,g) / 2 - b

    # Negative values are set to zero.
    cv2.threshold(src=R, dst=R, thresh=0., maxval=0.0, type=cv2.THRESH_TOZERO)
    cv2.threshold(src=G, dst=G, thresh=0., maxval=0.0, type=cv2.THRESH_TOZERO)
    cv2.threshold(src=B, dst=B, thresh=0., maxval=0.0, type=cv2.THRESH_TOZERO)
    cv2.threshold(src=Y, dst=Y, thresh=0., maxval=0.0, type=cv2.THRESH_TOZERO)

    image = cv2.merge((R,G,B,Y))
    return image

def diffImg(t0, t1, t2):
    d1 = cv2.absdiff(t2, t1)
    d2 = cv2.absdiff(t1, t0)
    return cv2.bitwise_and(d1, d2)


#Form Config

def are_similar(self, first, second):
        res = cv2.absdiff(first, second)
        hist = cv2.calcHist([res], [0], None, [256], [0, 256])
        return 1 - np.sum(hist[15::]) / np.sum(hist)

def frame_diff(prev_frame, cur_frame, next_frame):
    # Difference between the current frame and the next frame
    diff_frames_1 = cv2.absdiff(next_frame, cur_frame)

    # Difference between the current frame and the previous frame
    diff_frames_2 = cv2.absdiff(cur_frame, prev_frame)

    return cv2.bitwise_and(diff_frames_1, diff_frames_2)

# Define a function to get the current frame from the webcam

def have_motion(frame1, frame2):
    if frame1 is None or frame2 is None:
        return False
    delta = cv2.absdiff(frame1, frame2)
    thresh = cv2.threshold(delta, 25, 255, cv2.THRESH_BINARY)[1]
    return numpy.sum(thresh) > 0

def bgSubtract(rgb):
    return cv2.threshold(cv2.cvtColor(cv2.absdiff(rgb, bg), cv2.COLOR_BGR2GRAY), 32, 1, cv2.THRESH_BINARY)[1]

def __call__(self, image: numpy.ndarray, test):
        from scipy import stats

        def dilate_diff(image, range, iterations=1):
            dil = cv2.dilate(image, numpy.ones((range, range), numpy.float32), iterations=iterations)
            image = cv2.absdiff(image, dil)
            return image

        dtype = image.dtype
        rgb = (image.transpose(1, 2, 0) + 1) / 2
        lab = rgb2lab(rgb) / 100

        image = lab[:, :, 0]
        image = dilate_diff(image, 3).astype(numpy.float32)

        rand = 0.2 + (numpy.random.randn(1) / 20 if not test else 0)
        rand = 0.000001 if rand <= 0 else rand
        image = cv2.GaussianBlur(image, (5, 5), rand)

        rand = 0.4 + (numpy.random.randn(1) / 20 if not test else 0)
        rand = 0.000001 if rand <= 0 else rand
        image = cv2.GaussianBlur(image, (5, 5), rand)

        rand = numpy.random.randn(1) / 40 if not test else 0
        image = numpy.power(image, 0.8 + rand)

        image = image.astype(dtype)[numpy.newaxis]
        return image

def diff_rect(img1, img2, pos=None):
    """find counters include pos in differences between img1 & img2 (cv2 images)"""
    diff = cv2.absdiff(img1, img2)
    diff = cv2.GaussianBlur(diff, (3, 3), 0)
    edges = cv2.Canny(diff, 100, 200)
    _, thresh = cv2.threshold(edges, 0, 255, cv2.THRESH_BINARY)
    contours, _ = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_NONE)
    if not contours:
        return None
    contours.sort(key=lambda c: len(c))
    # no pos provide, just return the largest different area rect
    if pos is None:
        cnt = contours[-1]
        x0, y0, w, h = cv2.boundingRect(cnt)
        x1, y1 = x0+w, y0+h
        return (x0, y0, x1, y1)
    # else the rect should contain the pos
    x, y = pos
    for i in range(len(contours)):
        cnt = contours[-1-i]
        x0, y0, w, h = cv2.boundingRect(cnt)
        x1, y1 = x0+w, y0+h
        if x0 <= x <= x1 and y0 <= y <= y1:
            return (x0, y0, x1, y1)

def delta_images(t0, t1, t2):
    d1 = cv2.absdiff(t2, t0)
    return d1

def delta_images(t0, t1, t2):
    d1 = cv2.absdiff(t2, t0)
    return d1

def delta_images(t0, t1, t2):
    d1 = cv2.absdiff(t2, t0)
    return d1

def delta_images(t0, t1, t2):
    d1 = cv2.absdiff(t2, t0)
    return d1

def diffImg(t0,t1,t2):
    d1 = cv2.absdiff(t2,t1)
    d2 = cv2.absdiff(t1,t0)
    return cv2.bitwise_and(d1,d2)

def subtract_back(self,frm):
        #dst=self.__back__-self.__foreground__
        temp=np.zeros((600,800),np.uint8)

        self.__foreground__=cv2.blur(self.__foreground__,(3,3))
        dst=cv2.absdiff(self.__back__,self.__foreground__)

        #dst=cv2.adaptiveThreshold(dst,255,cv.CV_THRESH_BINARY,cv.CV_ADAPTIVE_THRESH_GAUSSIAN_C,5,10)
        val,dst=cv2.threshold(dst,0,255,cv.CV_THRESH_BINARY+cv.CV_THRESH_OTSU)

        fg=cv2.erode(dst,None,iterations=1)
        bg=cv2.dilate(dst,None,iterations=4)

        _,bg=cv2.threshold(bg,1,128,1)

        mark=cv2.add(fg,bg)
        mark32=np.int32(mark)
        #dst.copy(temp)

        #seq=cv.FindContours(cv.fromarray(dst),self.mem,cv.CV_RETR_EXTERNAL,cv.CV_CHAIN_APPROX_SIMPLE)
        #cntr,h=cv2.findContours(dst,cv.CV_RETR_EXTERNAL,cv.CV_CHAIN_APPROX_SIMPLE)
        #print cntr,h
        #cv.DrawContours(cv.fromarray(temp),seq,(255,255,255),(255,255,255),1,cv.CV_FILLED)
        cv2.watershed(frm, mark32)
        self.final_mask=cv2.convertScaleAbs(mark32)
        #print temp

        #--outputs---
        #cv2.imshow("subtraction",fg)
        #cv2.imshow("thres",dst)
        #cv2.imshow("thres1",bg)
        #cv2.imshow("mark",mark)
        #cv2.imshow("final",self.final_mask)

def __threshold_moving(input, last_image):
        """Thresholds off parts of the image that have moved or changed between
           the previous and next image.
        Args:
            input: A numpy.ndarray.
            last_image: The previous value of the numpy.ndarray.
        Returns:
            A numpy.ndarray with the parts that are the same in black.
        """
        if (last_image.shape == input.shape):
            output =  cv2.absdiff(input, last_image)
        else:
            output = numpy.ndarray(shape=input.shape, dtype=input.dtype)
        return input, output

def trackPoint(grayimage1, grayimage2):
    moveData = []   # initialize list of movementCenterPoints
    biggestArea = MIN_AREA
    # Get differences between the two greyed images
    differenceImage = cv2.absdiff( grayimage1, grayimage2 )
    # Blur difference image to enhance motion vectors
    differenceImage = cv2.blur( differenceImage,(BLUR_SIZE,BLUR_SIZE ))
    # Get threshold of blurred difference image based on THRESHOLD_SENSITIVITY variable
    retval, thresholdImage = cv2.threshold( differenceImage, THRESHOLD_SENSITIVITY, 255, cv2.THRESH_BINARY )
    try:
        thresholdImage, contours, hierarchy = cv2.findContours( thresholdImage, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE )
    except:
        contours, hierarchy = cv2.findContours( thresholdImage, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE )

    if contours != ():
        for c in contours:
            cArea = cv2.contourArea(c)
            if cArea > biggestArea:
                biggestArea = cArea
                ( x, y, w, h ) = cv2.boundingRect(c)
                cx = int(x + w/2)   # x center point of contour
                cy = int(y + h/2)   # y center point of contour
                moveData = [cx, cy, w, h]
    return moveData

#-----------------------------------------------------------------------------------------------

def detect(self, image):
        gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
        gray_image = cv2.equalizeHist(gray_image)
        blurred = cv2.GaussianBlur(gray_image, self.kernel, self.sigma)

        if self.prevImage is None:
            self.prevImage = blurred
        diff = cv2.absdiff(self.prevImage, blurred)
        _, binary = cv2.threshold(diff, 21, 255, cv2.THRESH_BINARY)

        if eval(cv2.__version__.split('.')[0]) == 3:
            _, cnts, hier = cv2.findContours(binary, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
        else:
            cnts, hier = cv2.findContours(binary, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

        cnts = sorted(cnts, key=cv2.contourArea, reverse=True)

        if len(cnts) < 1:
            is_detected = False
            contour = None
        else:
            largest_contour = cnts[0]
            if cv2.contourArea(largest_contour) < self.min_detection_area:
                is_detected = False
                contour = None
            else:
                is_detected = True
                contour = largest_contour
                self.prevImage = blurred
        return is_detected, contour