Python cv2.TM_SQDIFF Examples

def findAllMatches(self, needle, similarity):
        """ Find all matches for ``needle`` with confidence better than or equal to ``similarity``.

        Returns an array of tuples ``(position, confidence)`` if match(es) is/are found,
        or an empty array otherwise.
        """
        positions = []
        method = cv2.TM_CCOEFF_NORMED

        match = cv2.matchTemplate(self.haystack, self.needle, method)

        indices = (-match).argpartition(100, axis=None)[:100] # Review the 100 top matches
        unraveled_indices = numpy.array(numpy.unravel_index(indices, match.shape)).T
        for location in unraveled_indices:
            y, x = location
            confidence = match[y][x]
            if method == cv2.TM_SQDIFF_NORMED or method == cv2.TM_SQDIFF:
                if confidence <= 1-similarity:
                    positions.append(((x, y), confidence))
            else:
                if confidence >= similarity:
                    positions.append(((x, y), confidence))

        positions.sort(key=lambda x: (x[0][1], x[0][0]))
        return positions 

def locate_img(image, template):
    img = image.copy()
    res = cv2.matchTemplate(img, template, method)
    print res
    print res.shape
    cv2.imwrite('image/shape.png', res)
    min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)
    print cv2.minMaxLoc(res)
    if method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:
        top_left = min_loc
    else:
        top_left = max_loc
    h, w = template.shape
    bottom_right = (top_left[0] + w, top_left[1]+h)
    cv2.rectangle(img, top_left, bottom_right, 255, 2)
    cv2.imwrite('image/tt.jpg', img) 

def locate_img(image, template):
    img = image.copy()
    res = cv2.matchTemplate(img, template, method)
    print res
    print res.shape
    cv2.imwrite('image/shape.png', res)
    min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)
    print cv2.minMaxLoc(res)
    if method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:
        top_left = min_loc
    else:
        top_left = max_loc
    h, w = template.shape
    bottom_right = (top_left[0] + w, top_left[1]+h)
    cv2.rectangle(img, top_left, bottom_right, 255, 2)
    cv2.imwrite('image/tt.jpg', img) 

def find_watermark_from_gray(self, gray_img, watermark_template_gray_img):
        """
        从原图的灰度图中寻找水印位置
        :param gray_img: 原图的灰度图
        :param watermark_template_gray_img: 水印模板的灰度图
        :return: x1, y1, x2, y2
        """
        # Load the images in gray scale

        method = cv2.TM_CCOEFF
        # Apply template Matching
        res = cv2.matchTemplate(gray_img, watermark_template_gray_img, method)
        min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)

        # If the method is TM_SQDIFF or TM_SQDIFF_NORMED, take minimum
        if method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:
            x, y = min_loc
        else:
            x, y = max_loc

        return x, y, x + self.watermark_template_w, y + self.watermark_template_h 

def _run_match_template_on_rect(image, prev_image, rect, increased_rect):
    subimage = _get_subimage(image, increased_rect)
    prev_template = _get_subimage(prev_image, rect)

    match = cv2.matchTemplate(subimage, prev_template, cv2.TM_SQDIFF)

    min_val, max_val, min_loc, _ = cv2.minMaxLoc(match)

    dx, dy = min_loc
    template_h, template_w = prev_template.shape[:2]
    subimage_h, subimage_w = subimage.shape[:2]

    v_x = -(subimage_w / 2.) + dx + template_w / 2.
    v_y = -(subimage_h / 2.) + dy + template_h / 2.

    v = Point(v_x, v_y)

    _draw_match(match, min_val, max_val)

    return v 

def match_template1(template, img, plot=False, method=cv2.TM_SQDIFF_NORMED):
    img = cv2.imread(img, 0).copy()
    template = cv2.imread(template, 0)
    w, h = template.shape[::-1]
    if lib == OPENCV:
        res = cv2.matchTemplate(img, template, method)
        min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)
        if method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:
            top_left = min_loc
        else:
            top_left = max_loc
    else:
        result = match_template(img, template)
        ij = np.unravel_index(np.argmax(result), result.shape)
        top_left = ij[::-1]

    bottom_right = (top_left[0] + w, top_left[1] + h)

    if plot:
        cv2.rectangle(img, top_left, bottom_right, 255, 5)
        plt.subplot(121)
        plt.imshow(img)
        plt.title('Detected Point'), plt.xticks([]), plt.yticks([])
        plt.subplot(122)
        plt.imshow(template)

        plt.show()

    return top_left, bottom_right 

def MatchingMethod(param):

   global match_method
   match_method = param

   ## [copy_source]
   img_display = img.copy()
   ## [copy_source]
   ## [match_template]
   method_accepts_mask = (cv2.TM_SQDIFF == match_method or match_method == cv2.TM_CCORR_NORMED)
   if (use_mask and method_accepts_mask):
       result = cv2.matchTemplate(img, templ, match_method, None, mask)
   else:
       result = cv2.matchTemplate(img, templ, match_method)
   ## [match_template]

   ## [normalize]
   cv2.normalize( result, result, 0, 1, cv2.NORM_MINMAX, -1 )
   ## [normalize]
   ## [best_match]
   minVal, maxVal, minLoc, maxLoc = cv2.minMaxLoc(result, None)
   ## [best_match]

   ## [match_loc]
   if (match_method == cv2.TM_SQDIFF or match_method == cv2.TM_SQDIFF_NORMED):
       matchLoc = minLoc
   else:
       matchLoc = maxLoc
   ## [match_loc]

   ## [imshow]
   cv2.rectangle(img_display, matchLoc, (matchLoc[0] + templ.shape[0], matchLoc[1] + templ.shape[1]), (0,0,0), 2, 8, 0 )
   cv2.rectangle(result, matchLoc, (matchLoc[0] + templ.shape[0], matchLoc[1] + templ.shape[1]), (0,0,0), 2, 8, 0 )
   cv2.imshow(image_window, img_display)
   cv2.imshow(result_window, result)
   ## [imshow]
   pass 

def findBestMatch(self, needle, similarity):
        """ Find the best match for ``needle`` that has a similarity better than or equal to ``similarity``.

        Returns a tuple of ``(position, confidence)`` if a match is found, or ``None`` otherwise.

        *Developer's Note - Despite the name, this method actually returns the **first** result
        with enough similarity, not the **best** result.*
        """
        method = cv2.TM_CCOEFF_NORMED
        position = None

        match = cv2.matchTemplate(self.haystack, needle, method)
        min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(match)
        if method == cv2.TM_SQDIFF_NORMED or method == cv2.TM_SQDIFF:
            confidence = min_val
            if min_val <= 1-similarity:
                # Confidence checks out
                position = min_loc
        else:
            confidence = max_val
            if max_val >= similarity:
                # Confidence checks out
                position = max_loc

        if not position:
            return None

        return (position, confidence) 

def search(img):
    result = cv2.matchTemplate(img, template, cv2.TM_SQDIFF)
    min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(result)

    cv2.rectangle(img, (min_loc[0], min_loc[1]), (min_loc[0] + template_size[1], min_loc[1] + template_size[0]), (255, 0, 0), 4)

    return img, min_loc[0] + template_size[1] / 2, min_loc[1] +  template_size[0] 

def getRefCoordinate(self, image, template):
#        method = cv2.TM_SQDIFF                     #2
        method = cv2.TM_SQDIFF_NORMED              #1
#        method = cv2.TM_CCORR_NORMED                #3
        method = cv2.TM_CCOEFF_NORMED                #4
        res = cv2.matchTemplate(image, template, method)
        min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)
        # If the method is TM_SQDIFF or TM_SQDIFF_NORMED, take minimum
        if method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:
            top_left = min_loc
        else:
            top_left = max_loc
#        bottom_right = (top_left[0] + w, top_left[1] + h)
        return top_left 

def _find_in_image_avg_diff(cls, find_image, in_image):
        res = cv2.matchTemplate(in_image, find_image, cv2.TM_SQDIFF)
        min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)

        top_left = min_loc
        bottom_right = (top_left[0] + find_image.shape[1],
                        top_left[1] + find_image.shape[0])
        image_found = in_image[top_left[1]:bottom_right[1],
                               top_left[0]:bottom_right[0],
                               :]
        diff = np.abs(image_found.astype(np.float32)
                      - find_image.astype(np.float32))
        return np.average(diff) 

def find(search_file, image_file, threshold=0.7):
    '''
    Locate image position with cv2.templateFind

    Use pixel match to find pictures.

    Args:
        search_file(string): filename of search object
        image_file(string): filename of image to search on
        threshold: optional variable, to ensure the match rate should >= threshold

    Returns:
        A tuple like (x, y) or None if nothing found

    Raises:
        IOError: when file read error
    '''
    search = _cv2open(search_file)
    image  = _cv2open(image_file)

    w, h = search.shape[::-1]

    method = cv2.CV_TM_CCORR_NORMED
    res = cv2.matchTemplate(image, search, method)
    min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)
    if method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:
        top_left = min_loc
    else:
        top_left = max_loc
    bottom_right = (top_left[0] + w, top_left[1] + h)
    middle_point = (top_left[0]+w/2, top_left[1]+h/2)
    print top_left, bottom_right
    return middle_point

    # if len(region_center):
    #     x = int(maxloc[0]+region_center[0]-source_width/2)
    #     y = int(maxloc[1]+region_center[1]-source_height/2)
    # else:
    #     [x,y] = maxloc
    # return max_val, [x,y] 

def search(img):
    result = cv2.matchTemplate(img, template, cv2.TM_SQDIFF)
    min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(result)

    cv2.rectangle(
        img,
        (min_loc[0], min_loc[1]),
        (min_loc[0] + template_size[1], min_loc[1] + template_size[0]),
        (255, 0, 0),
        4)
    return img, min_loc[0] + template_size[1] / 2, min_loc[1] +  template_size[0] 

def match_template(self,img,template,threshold=u"0.8"):
        """  Matches a template in an image using TM_CCOEFF_NORMED method

        Both `img` and `tempalte` are BGR ndarray object.
        The result is in the the center and boundary of the match.
        """
        _method = cv2.TM_CCOEFF_NORMED
        gray_img = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
        gray_template = cv2.cvtColor(template,cv2.COLOR_BGR2GRAY)
        w,h = gray_template.shape[::-1]

        res = cv2.matchTemplate(gray_img,gray_template,_method)
        loc = np.where(res >= float(threshold))
        if len(loc[0]) != 0 and len(loc[1]) != 0:
            min_val,max_val,min_loc,max_loc = cv2.minMaxLoc(res)
            if _method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:
                top_left = min_loc
            else:
                top_left = max_loc
            bottom_right = (top_left[0] + w, top_left[1] + h)
            mx = int((top_left[0] + bottom_right[0])/2)
            my = int((top_left[1] + bottom_right[1])/2)
            result = ((mx,my),(top_left[0],top_left[1],bottom_right[0],bottom_right[1]))
            BuiltIn().log("Found image at %s" % str(result))
        else:
            result = (None,None)
            BuiltIn().log("WRN: Could not found the template")
        return result 

def pokemon_image_matching(pokemon_image_name, fort_img_name, is_pokemon):

    pokemon_image = cv2.imread(pokemon_image_name, cv2.IMREAD_UNCHANGED)
    fort_img = cv2.imread(fort_img_name, 3)

    if pokemon_image is None or fort_img is None:
        return 100000.0

    croped = pokemon_image[0:256,0:190]

    height_f, width_f, channels_f = fort_img.shape
    scale = 147 / 256 * width_f / 133

    scaled = cv2.resize(croped, None, fx=scale, fy=scale)

    scaled_h, scaled_w, scaled_c = scaled.shape
    channels = cv2.split(scaled)

    if is_pokemon:
        scale_crop_fort = width_f / 156
        target_x = (16*scale_crop_fort)
        target_y = (28*scale_crop_fort)
        fort_img = fort_img[target_x-2:target_x+2+scaled_h, target_y-2:target_y+2+scaled_w]
    else:
        scale_crop_fort = width_f / 133
        target_x = int(12*scale_crop_fort)
        target_y = int(24*scale_crop_fort)
        fort_img = fort_img[target_x-2:target_x+2+scaled_h, target_y-2:target_y+2+scaled_w]

    scaled_no_alpth = cv2.merge([channels[0], channels[1], channels[2]])
    transparent_mask = cv2.merge([channels[3], channels[3], channels[3]])

    white_pixels = channels[3].sum()/255

    result = cv2.matchTemplate(fort_img, scaled_no_alpth, cv2.TM_SQDIFF, mask=transparent_mask)

    min_val3, max_val3, min_loc3, max_loc3 = cv2.minMaxLoc(result)

    min_val3 = min_val3 / white_pixels

    return min_val3