Python cv2.ADAPTIVE_THRESH_MEAN_C Examples

def getSignature(img):
    imgSize = np.shape(img)

    gImg = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

    # Adaptive Thresholding requires the blocksize to be odd and bigger than 1
    blockSize = 1 / 8 * imgSize[0] / 2 * 2 + 1
    if blockSize <= 1:
        blockSize = imgSize[0] / 2 * 2 + 1
    const = 10

    mask = cv2.adaptiveThreshold(gImg, maxValue = 255, adaptiveMethod = cv2.ADAPTIVE_THRESH_MEAN_C, thresholdType = cv2.THRESH_BINARY, blockSize = blockSize, C = const)
    rmask = cv2.bitwise_not(mask)

    return (cv2.bitwise_and(img, img, mask=rmask), rmask)

# First Prompt 

def get_name(img):
        #    cv2.imshow("method3", img)
        #    cv2.waitKey()
        print('name')
        _, _, red = cv2.split(img) #split 会自动将UMat转换回Mat
        red = cv2.UMat(red)
        red = hist_equal(red)
        red = cv2.adaptiveThreshold(red, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 151, 50)
        #    red = cv2.medianBlur(red, 3)
        red = img_resize(red, 150)
        img = img_resize(img, 150)
        # showimg(red)
        # cv2.imwrite('name.png', red)
        #    img2 = Image.open('address.png')
        # img = Image.fromarray(cv2.UMat.get(red).astype('uint8'))
        #return get_result_vary_length(red, 'chi_sim', img, '-psm 7')
        return get_result_vary_length(red, 'chi_sim', img, '--psm 7')
        # return punc_filter(pytesseract.image_to_string(img, lang='chi_sim', config='-psm 13').replace(" ","")) 

def adaptiveThreshold(plates):
    for i, plate in enumerate(plates):
        img = cv2.imread(plate)

        gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
        cv2.imshow('gray', gray)

        ret, thresh = cv2.threshold(gray, 50, 255, cv2.THRESH_BINARY)
        # cv2.imshow('thresh', thresh)

        threshMean = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 5, 10)
        # cv2.imshow('threshMean', threshMean)

        threshGauss = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 51, 27)
        cv2.imshow('threshGauss', threshGauss)
        cv2.imwrite("processed\\plate{}.png".format(i), threshGauss)

        cv2.waitKey(0) 

def read_file(fname):
    image = cv2.imread(fname,0)


    image = cv2.adaptiveThreshold(image,255,cv2.ADAPTIVE_THRESH_MEAN_C,cv2.THRESH_BINARY,11,2)
    # cv2.namedWindow('image', cv2.WINDOW_NORMAL)
    # cv2.imshow('image',image)
    # cv2.waitKey(0)
    # cv2.destroyAllWindows()

    # image = cv2.adaptiveThreshold(image,255,cv2.ADAPTIVE_THRESH_GAUSSIAN_C,cv2.THRESH_BINARY,11,2)
    # cv2.imwrite("/home/ggdhines/temp.jpg",image)
    # assert False


    # _,image = cv2.threshold(image,200,255,cv2.THRESH_BINARY)

    # image = 255 - image
    # image = image > 0
    image = image.astype(np.float)

    return image 

def update_image(image_id, category_id = 0, image_filenames=[], enable_vis=True, enable_marker_dump=False):
    try:
        global contours, hierarchy, img, gray, g_image_filenames
        if len(image_filenames) > 0:
            g_image_filenames=image_filenames
        img=cv.imread(g_image_filenames[image_id])
        # print(g_image_filenames[image_id])
        cv.setTrackbarPos('image', 'marker', image_id)

        gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
        gray[np.where(gray <= [3])] = [187]
        gray = cv.medianBlur(gray, 11)

        if enable_vis:
            cv.imshow('gray', gray)

        if CANNY_MODE:
            thrs1 = cv.getTrackbarPos('thrs1', 'marker')
            thrs2 = cv.getTrackbarPos('thrs2', 'marker')
            bin = cv.Canny(gray, thrs1, thrs2, apertureSize=5)
        else:
            bin = cv.adaptiveThreshold(
                gray, 255, cv.ADAPTIVE_THRESH_MEAN_C, cv.THRESH_BINARY_INV, 31, 10)

        if enable_vis:
            cv.imshow('bin', bin)

        _, contours0, hierarchy = cv.findContours(
            bin.copy(), cv.RETR_EXTERNAL, cv.CHAIN_APPROX_SIMPLE)
        contours = [cnt for cnt in contours0 if cv.contourArea(cnt) > 200]

        if enable_vis:
            cv.imshow('image', img)
        update_contour(category_id, image_id, enable_vis, enable_marker_dump)
    except Exception:
        import traceback
        traceback.print_exc()
        raise 

def remove_noise_and_smooth(file_name):
    logging.info('Removing noise and smoothening image')
    img = cv2.imread(file_name, 0)
    filtered = cv2.adaptiveThreshold(img.astype(np.uint8), 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 41, 3)
    kernel = np.ones((1, 1), np.uint8)
    opening = cv2.morphologyEx(filtered, cv2.MORPH_OPEN, kernel)
    closing = cv2.morphologyEx(opening, cv2.MORPH_CLOSE, kernel)
    img = image_smoothening(img)
    or_image = cv2.bitwise_or(img, closing)
    return or_image 

def binarization(image, block_size=5, C=10):
        """
        convert input image to binary image
        cv2.adaptiveThreshold is used, for detailed information, refer to opencv docs
        :param image:
        :return:
        """
        if image.ndim == 3:
            image_grayscale = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
        else:
            image_grayscale = image

        binary_image = cv2.adaptiveThreshold(image_grayscale, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
                                             cv2.THRESH_BINARY, block_size, C)
        return binary_image 

def processing(self, img):
        super().processing(img)
        print("真正的核心算法:边缘提取算法")
        newImg = cv2.adaptiveThreshold(img, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 5, 10)
        return newImg 

def mean(gray_img, max_value, object_type="light"):
    """Creates a binary image from a grayscale image based on the mean adaptive threshold method.

    Inputs:
    gray_img     = Grayscale image data
    max_value    = value to apply above threshold (usually 255 = white)
    object_type  = "light" or "dark" (default: "light")
                   - If object is lighter than the background then standard thresholding is done
                   - If object is darker than the background then inverse thresholding is done

    Returns:
    bin_img      = Thresholded, binary image

    :param gray_img: numpy.ndarray
    :param max_value: int
    :param object_type: str
    :return bin_img: numpy.ndarray
    """
    # Set the threshold method
    threshold_method = ""
    if object_type.upper() == "LIGHT":
        threshold_method = cv2.THRESH_BINARY
    elif object_type.upper() == "DARK":
        threshold_method = cv2.THRESH_BINARY_INV
    else:
        fatal_error('Object type ' + str(object_type) + ' is not "light" or "dark"!')

    params.device += 1

    bin_img = _call_adaptive_threshold(gray_img, max_value, cv2.ADAPTIVE_THRESH_MEAN_C, threshold_method,
                                       "_mean_threshold_")

    return bin_img


# Otsu autothreshold 

def main():
    const = 2
    block_size = 51
    imagePath = "../data/sudoku.png"

    image = cv2.imread(imagePath, 0)

    cv2.imshow("Orignal Image", image)

    mean_thresh = \
        cv2.adaptiveThreshold(image, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, block_size, const)
    gaussian_thresh = \
        cv2.adaptiveThreshold(image, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, block_size, const)

    images = [image, mean_thresh, gaussian_thresh]
    titles = ["Orignals", "Mean", "Gaussian"]

    cv2.imshow("Mean Image", mean_thresh)
    cv2.imshow("Gaussian Image", gaussian_thresh)

    for i in range(3):
        plt.subplot(3, 1, i + 1)
        plt.imshow(images[i], cmap='gray')
        plt.title(titles[i])

    plt.show()

    cv2.waitKey(0)
    cv2.destroyAllWindows() 

def thresholdify(self, img):
        img = cv2.adaptiveThreshold(img.astype(np.uint8), 255, cv2.ADAPTIVE_THRESH_MEAN_C,
                                    cv2.THRESH_BINARY, 11, 3)
        return 255 - img 

def binarization(image, block_size=5, C=10):
        """
        convert input image to binary image
        cv2.adaptiveThreshold is used, for detailed information, refer to opencv docs
        :param image:
        :return:
        """
        if image.ndim == 3:
            image_grayscale = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
        else:
            image_grayscale = image

        binary_image = cv2.adaptiveThreshold(image_grayscale, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
                                             cv2.THRESH_BINARY, block_size, C)
        return binary_image 

def preprocess_image(self):

        gray = self.image

        #Applying Gaussian Blur to smooth out the noise
        gray = cv2.GaussianBlur(gray, (11, 11), 0)
        try:
            os.remove("StagesImages/1.jpg")
        except:
            pass
        cv2.imwrite("StagesImages/1.jpg", gray)

        # Applying thresholding using adaptive Gaussian|Mean thresholding
        gray = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C | cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 5, 2)
        try:
            os.remove("StagesImages/2.jpg")
        except:
            pass
        cv2.imwrite("StagesImages/2.jpg", gray)

        #Inverting the image
        gray = cv2.bitwise_not(gray)
        try:
            os.remove("StagesImages/3.jpg")
        except:
            pass
        cv2.imwrite("StagesImages/3.jpg", gray)

        #Dilating the image to fill up the "cracks" in lines
        kernel = np.array([[0, 1, 0], [1, 1, 1], [0, 1, 0]], np.uint8)
        gray = cv2.dilate(gray, kernel)
        self.image = gray
        try:
            os.remove("StagesImages/4.jpg")
        except:
            pass
        cv2.imwrite("StagesImages/4.jpg", gray) 

def draw_markers(img, workspace_ratio=1.0):
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

    img_thresh = cv2.adaptiveThreshold(gray, maxValue=255, adaptiveMethod=cv2.ADAPTIVE_THRESH_MEAN_C,
                                       thresholdType=cv2.THRESH_BINARY, blockSize=15, C=32)

    list_good_candidates = find_markers_from_img_thresh(img_thresh)
    if not list_good_candidates:
        return False, img
    im_draw = img.copy()
    for marker in list_good_candidates:
        cx, cy = marker.get_center()
        radius = marker.get_radius()
        cv2.circle(im_draw, (cx, cy), radius, (0, 0, 255), 2)
    if len(list_good_candidates) > 6:
        return False, im_draw

    if len(list_good_candidates) == 4:
        list_good_candidates = sort_markers_detection(list_good_candidates)
    else:
        list_good_candidates = complicated_sort_markers(list_good_candidates, workspace_ratio=workspace_ratio)
        if list_good_candidates is None:
            return False, im_draw

    for i, marker in enumerate(list_good_candidates[:4]):
        cx, cy = marker.get_center()
        radius = marker.get_radius()
        cv2.circle(im_draw, (cx, cy), radius, (0, 200, 0), 2)
        cv2.putText(im_draw, "{}".format(i + 1),
                    (cx + 5, cy - 15), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 0, 0), 3)
        cv2.putText(im_draw, "{}".format(i + 1),
                    (cx + 5, cy - 15), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 200, 0), 2)
    return True, im_draw 

def extract_img_markers(img, workspace_ratio=1.0):
    """
    Extract working area from an image thanks to 4 Niryo's markers
    :param img: OpenCV image which contain 4 Niryo's markers
    :param workspace_ratio: Ratio between the width and the height of the area represented by the markers
    :return: extracted and warped working area image
    """

    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

    img_thresh = cv2.adaptiveThreshold(gray, maxValue=255, adaptiveMethod=cv2.ADAPTIVE_THRESH_MEAN_C,
                                       thresholdType=cv2.THRESH_BINARY, blockSize=15, C=25)

    list_good_candidates = find_markers_from_img_thresh(img_thresh)
    if not list_good_candidates or len(list_good_candidates) > 6:
        return None

    if len(list_good_candidates) == 4:
        list_good_candidates = sort_markers_detection(list_good_candidates)
    else:
        list_good_candidates = complicated_sort_markers(list_good_candidates, workspace_ratio=workspace_ratio)
        if list_good_candidates is None:
            return None

    im_cut = extract_sub_img(img, list_good_candidates, ratio_w_h=workspace_ratio)
    return im_cut 

def gradient_and_binary(img_blurred, image_name='1.jpg', save_path='./'):  # 将灰度图二值化，后面两个参数调试用
    """
    求取梯度，二值化
    :param img_blurred: 滤波后的图片
    :param image_name: 图片名，测试用
    :param save_path: 保存路径，测试用
    :return:  二值化后的图片
    """
    gradX = cv2.Sobel(img_blurred, ddepth=cv2.CV_32F, dx=1, dy=0)
    gradY = cv2.Sobel(img_blurred, ddepth=cv2.CV_32F, dx=0, dy=1)
    img_gradient = cv2.subtract(gradX, gradY)
    img_gradient = cv2.convertScaleAbs(img_gradient)  # sobel算子,计算梯度, 也可以用canny算子替代

    # 这里改进成自适应阈值,貌似没用
    img_thresh = cv2.adaptiveThreshold(img_gradient, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 3, -3)
    # cv2.imwrite(os.path.join(save_path, img_name + '_binary.jpg'), img_thresh)  # 二值化 阈值未调整好

    kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (5, 5))
    img_closed = cv2.morphologyEx(img_thresh, cv2.MORPH_CLOSE, kernel)
    img_closed = cv2.morphologyEx(img_closed, cv2.MORPH_OPEN, kernel)
    img_closed = cv2.erode(img_closed, None, iterations=9)
    img_closed = cv2.dilate(img_closed, None, iterations=9)  # 腐蚀膨胀
    # 这里调整了kernel大小(减小),腐蚀膨胀次数后(增大),出错的概率大幅减小

    return img_closed 

def getSignature(img):
    imgSize = np.shape(img)

    gImg = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

    # minBlockSize = 3
    # maxBlockSize = 101
    # minC = 3
    # maxC = 101
    #
    # bestContourNo = 1000000
    # bestBlockSize = 0
    # bestC = 0
    #
    # for c in range(minC, maxC, 2):
    #     for bs in range(minBlockSize, maxBlockSize, 2):
    #         mask = cv2.adaptiveThreshold(gImg, maxValue = 255, adaptiveMethod = cv2.ADAPTIVE_THRESH_MEAN_C, thresholdType = cv2.THRESH_BINARY, blockSize = bs, C = c)
    #         rmask = cv2.bitwise_not(mask)
    #         _, contours, _ = cv2.findContours(image = rmask.copy(), mode = cv2.RETR_TREE, method = cv2.CHAIN_APPROX_SIMPLE)
    #         if len(contours) > 15 and len(contours) < bestContourNo:
    #             bestContourNo = len(contours)
    #             bestBlockSize = bs
    #             bestC = c

    # blockSize = 21, C = 10

    # TODO throw error if blockSize is bigger than image
    blockSize = 21
    C = 10
    if blockSize > imgSize[0]:
        if imgSize[0] % 2 == 0:
            blockSize = imgSize[0] - 1
        else:
            blockSize = imgSize[0]

    if blockSize > imgSize[1]:
        if imgSize[0] % 2 == 0:
            blockSize = imgSize[1] - 1
        else:
            blockSize = imgSize[1]

    mask = cv2.adaptiveThreshold(gImg, maxValue = 255, adaptiveMethod = cv2.ADAPTIVE_THRESH_MEAN_C, thresholdType = cv2.THRESH_BINARY, blockSize = blockSize, C = C)
    rmask = cv2.bitwise_not(mask)

    return cv2.bitwise_and(signature, signature, mask=rmask)

# Camera capture
# camera = cv2.VideoCapture(0)
# (grabbed, signature) = camera.read()
# cv2.imshow('Picture', signature)
# cv2.waitKey(0) 

def filter_lane_points(self,
                           img,
                           filter_type='bilateral',
                           ksize_r=25,
                           C_r=8,
                           ksize_b=35,
                           C_b=5,
                           mask_noise=False,
                           ksize_noise=65,
                           C_noise=10,
                           noise_thresh=135):
        '''
        Filter an image to isolate lane lines and return a binary version.

        All image color space conversion, thresholding, filtering and morphing
        happens inside this method. It takes an RGB color image as input and
        returns a binary filtered version.
        '''

        # Define structuring elements for cv2 functions
        strel_lab_b = cv2.getStructuringElement(shape=cv2.MORPH_ELLIPSE, ksize=(55,55))
        strel_rgb_r = cv2.getStructuringElement(shape=cv2.MORPH_ELLIPSE, ksize=(29,29))
        strel_open = cv2.getStructuringElement(shape=cv2.MORPH_ELLIPSE, ksize=(5,5))
        # Extract RGB R-channel and LAB B-channel
        rgb_r_channel = img[:,:,0]
        lab_b_channel = (cv2.cvtColor(img, cv2.COLOR_RGB2LAB))[:,:,2]
        # Apply tophat morphology
        rgb_r_tophat = cv2.morphologyEx(rgb_r_channel, cv2.MORPH_TOPHAT, strel_rgb_r, iterations=1)
        lab_b_tophat = cv2.morphologyEx(lab_b_channel, cv2.MORPH_TOPHAT, strel_lab_b, iterations=1)
        if filter_type == 'bilateral':
            # Apply bilateral adaptive color thresholding
            rgb_r_thresh = bilateral_adaptive_threshold(rgb_r_tophat, ksize=ksize_r, C=C_r)
            lab_b_thresh = bilateral_adaptive_threshold(lab_b_tophat, ksize=ksize_b, C=C_b)
        elif filter_type == 'neighborhood':
            rgb_r_thresh = cv2.adaptiveThreshold(rgb_r_channel, 255, adaptiveMethod=cv2.ADAPTIVE_THRESH_MEAN_C, thresholdType=cv2.THRESH_BINARY, blockSize=ksize_r, C=-C_r)
            lab_b_thresh = cv2.adaptiveThreshold(lab_b_channel, 255, adaptiveMethod=cv2.ADAPTIVE_THRESH_MEAN_C, thresholdType=cv2.THRESH_BINARY, blockSize=ksize_b, C=-C_b)
        else:
            raise ValueError("Unexpected filter mode. Expected modes are 'bilateral' or 'neighborhood'.")
        if mask_noise: # Merge both color channels and the noise mask
            # Create a mask to filter out noise such as trees and other greenery based on the LAB B-channel
            noise_mask_part1 = cv2.inRange(lab_b_channel, noise_thresh, 255) # This catches the noise, but unfortunately also the yellow line, therefore...
            noise_mask_part2 = bilateral_adaptive_threshold(lab_b_channel, ksize=ksize_noise, C=C_noise) # ...this brings the yellow line back...
            noise_bool = np.logical_or(np.logical_not(noise_mask_part1), noise_mask_part2) # ...once we combine the two.
            noise_mask = np.zeros_like(rgb_r_channel, dtype=np.uint8)
            noise_mask[noise_bool] = 255

            merged_bool = np.logical_and(np.logical_or(rgb_r_thresh, lab_b_thresh), noise_mask)
            merged = np.zeros_like(rgb_r_channel, dtype=np.uint8)
            merged[merged_bool] = 255
        else: # Only merge the two color channels
            merged_bool = np.logical_or(rgb_r_thresh, lab_b_thresh)
            merged = np.zeros_like(rgb_r_channel, dtype=np.uint8)
            merged[merged_bool] = 255

        # Apply open morphology
        opened = cv2.morphologyEx(merged, cv2.MORPH_OPEN, strel_open, iterations=1)

        return opened 

def remap_image(name, img, small, page_dims, params):

    height = 0.5 * page_dims[1] * OUTPUT_ZOOM * img.shape[0]
    height = round_nearest_multiple(height, REMAP_DECIMATE)

    width = round_nearest_multiple(height * page_dims[0] / page_dims[1],
                                   REMAP_DECIMATE)

    print '  output will be {}x{}'.format(width, height)

    height_small = height / REMAP_DECIMATE
    width_small = width / REMAP_DECIMATE

    page_x_range = np.linspace(0, page_dims[0], width_small)
    page_y_range = np.linspace(0, page_dims[1], height_small)

    page_x_coords, page_y_coords = np.meshgrid(page_x_range, page_y_range)

    page_xy_coords = np.hstack((page_x_coords.flatten().reshape((-1, 1)),
                                page_y_coords.flatten().reshape((-1, 1))))

    page_xy_coords = page_xy_coords.astype(np.float32)

    image_points = project_xy(page_xy_coords, params)
    image_points = norm2pix(img.shape, image_points, False)

    image_x_coords = image_points[:, 0, 0].reshape(page_x_coords.shape)
    image_y_coords = image_points[:, 0, 1].reshape(page_y_coords.shape)

    image_x_coords = cv2.resize(image_x_coords, (width, height),
                                interpolation=cv2.INTER_CUBIC)

    image_y_coords = cv2.resize(image_y_coords, (width, height),
                                interpolation=cv2.INTER_CUBIC)

    img_gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)

    remapped = cv2.remap(img_gray, image_x_coords, image_y_coords,
                         cv2.INTER_CUBIC,
                         None, cv2.BORDER_REPLICATE)

    thresh = cv2.adaptiveThreshold(remapped, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
                                   cv2.THRESH_BINARY, ADAPTIVE_WINSZ, 25)

    pil_image = Image.fromarray(thresh)
    pil_image = pil_image.convert('1')

    threshfile = name + '_thresh.png'
    pil_image.save(threshfile, dpi=(OUTPUT_DPI, OUTPUT_DPI))

    if DEBUG_LEVEL >= 1:
        height = small.shape[0]
        width = int(round(height * float(thresh.shape[1])/thresh.shape[0]))
        display = cv2.resize(thresh, (width, height),
                             interpolation=cv2.INTER_AREA)
        debug_show(name, 6, 'output', display)

    return threshfile 

def get_mask(name, small, pagemask, masktype):

    sgray = cv2.cvtColor(small, cv2.COLOR_RGB2GRAY)

    if masktype == 'text':

        mask = cv2.adaptiveThreshold(sgray, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
                                     cv2.THRESH_BINARY_INV,
                                     ADAPTIVE_WINSZ,
                                     25)

        if DEBUG_LEVEL >= 3:
            debug_show(name, 0.1, 'thresholded', mask)

        mask = cv2.dilate(mask, box(9, 1))

        if DEBUG_LEVEL >= 3:
            debug_show(name, 0.2, 'dilated', mask)

        mask = cv2.erode(mask, box(1, 3))

        if DEBUG_LEVEL >= 3:
            debug_show(name, 0.3, 'eroded', mask)

    else:

        mask = cv2.adaptiveThreshold(sgray, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
                                     cv2.THRESH_BINARY_INV,
                                     ADAPTIVE_WINSZ,
                                     7)

        if DEBUG_LEVEL >= 3:
            debug_show(name, 0.4, 'thresholded', mask)

        mask = cv2.erode(mask, box(3, 1), iterations=3)

        if DEBUG_LEVEL >= 3:
            debug_show(name, 0.5, 'eroded', mask)

        mask = cv2.dilate(mask, box(8, 2))

        if DEBUG_LEVEL >= 3:
            debug_show(name, 0.6, 'dilated', mask)

    return np.minimum(mask, pagemask) 

def show_em(self):
        """Displays the EM events (grayscale ATIS events)"""
        frame_length = 24e3
        t_max = self.data.ts[-1]
        frame_start = self.data[0].ts
        frame_end = self.data[0].ts + frame_length
        max_val = 1.16e5
        min_val = 1.74e3
        val_range = max_val - min_val

        thr = np.rec.array(None, dtype=[('valid', np.bool_), ('low', np.uint64), ('high', np.uint64)], shape=(self.height, self.width))
        thr.valid.fill(False)
        thr.low.fill(frame_start)
        thr.high.fill(0)

        def show_em_frame(frame_data):
            """Prepare and show a single frame of em data to be shown"""
            for datum in np.nditer(frame_data):
                ts_val = datum['ts'].item(0)
                thr_data = thr[datum['y'].item(0), datum['x'].item(0)]

                if datum['p'].item(0) == 0:
                    thr_data.valid = 1
                    thr_data.low = ts_val
                elif thr_data.valid == 1:
                    thr_data.valid = 0
                    thr_data.high = ts_val - thr_data.low

            img = 255 * (1 - (thr.high - min_val) / (val_range))
            #thr_h = cv2.adaptiveThreshold(thr_h, 255,
            #cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 3, 0)
            img = np.piecewise(img, [img <= 0, (img > 0) & (img < 255), img >= 255], [0, lambda x: x, 255])
            img = img.astype('uint8')
            cv2.imshow('img', img)
            cv2.waitKey(1)

        while frame_start < t_max:
            #with timer.Timer() as em_playback_timer:
            frame_data = self.data[(self.data.ts >= frame_start) & (self.data.ts < frame_end)]
            show_em_frame(frame_data)
            frame_start = frame_end + 1
            frame_end += frame_length + 1
            #print 'showing em frame took %s seconds' %em_playback_timer.secs

        cv2.destroyAllWindows()
        return