pypocquant.lib.barcode

Module Contents

Classes

Barcode

Pythonic barcode object.

Functions

detect(image: np.ndarray, expected_area=22000, expected_aspect_ratio=7.5, barcode_border=75, blur_size=(3, 3), morph_rect=(9, 3), mm_iter=1, qc=True, verbose=False)

Detect the barcode in the image.

rotate(image, angle)

Rotate the image by given angle in degrees.

calc_area_and_approx_aspect_ratio(contour)

Calculate area and approximate aspect ratio of a contour.

rotate_90_if_needed(image)

Try to estimate the orientation of the image, and rotate if needed.

read_FID_from_barcode_image(image)

Read the FID string from the barcode image using pytesseract and

get_fid_from_barcode_data(barcode_data, barcode_type='CODE128')

Parse the output of pyzbar and retrieve the FID.

get_fid_from_box_image_using_ocr(box_img)

Use pytesseract to retrieve FID from the strip box image.

try_extracting_barcode_from_box_with_rotations(box, scaling=(1.0, 0.5, 0.25), verbose=False, log_list=None)

Try extracting barcode from QR code box while scaling it for different orientations [0, 90, 180, -90].

try_extracting_barcode_with_rotation(image, angle_range=15, verbose=True, log_list: list = None)

Try extracting barcode from QR code box for a list of angles in the range of angle_range.

find_strip_box_from_barcode_data_fh(image, barcode_data, qr_code_border=30, qc=False)

Extract the box around the strip using the QR barcode data.

find_strip_box_from_barcode_data(image, barcode_data, qr_code_border=30, qr_code_spacer=40, barcode_border=80, qc=False)

Extract the box around the strip using the QR barcode data.

try_extracting_barcode_with_linear_stretch(image, lower_bound_range=(25, ), upper_bound_range=(98, ))

Try to extract the barcodes from the image by rescaling the intensity of the image with a linear stretch.

try_getting_fid_from_code128_barcode(barcode_data)

Try finding a CODE 128 barcode in barcode data that should contain the patient FID.

try_get_fid_from_rgb(image)

Extract FID from rgb image.

try_extracting_fid_and_all_barcodes_with_linear_stretch_fh(image, lower_bound_range=(0, 5, 15, 25, 35), upper_bound_range=(100, 98, 95, 92, 89), scaling=(1.0, ))

Try extracting the fid and all barcodes from the image by rescaling the intensity of the image with a

try_extracting_all_barcodes_with_linear_stretch(image, lower_bound_range=(0, 5, 15, 25, 35), upper_bound_range=(100, 98, 95, 92, 89))

Try extracting the fid and all barcodes from the image by rescaling the intensity of the image with a

rotate_if_needed_fh(image, barcode_data, image_log, verbose=True)

Rotate the image if the orientation is not the expected one.

rotate_if_needed(image, barcode_data, image_log, verbose=True)

Rotate the image if the orientation is not the expected one.

pick_FID_from_candidates(fid_pyzbar, fid_tesseract)

Selection of FID from candidates depending on if candidates contain a FID.

mask_strip(strip_gray, x_barcode, qr_code_extents)

Hide the barcode on the strip image.

extract_strip_from_box(box, qr_code_width, qr_code_height, qr_code_spacer=40, slack=0)

Extract the strip from the strip box.

get_fid_numeric_value_fh(fid)

Return the numeric value of the FID (as string).

get_fid_numeric_value(fid)

Return the numeric value of the FID.

get_box_rotation_angle(pt1, pt2, pt3)

Determine the the QR code box rotation angle

align_box_with_image_border_fh(barcode_data, image)

Method to align QR code box with image border of the full image (old pipeline).

align_box_with_image_border(barcode_data, image)

Method to align QR code box with image border of the full image.
class pypocquant.lib.barcode.Barcode(top: int, left: int, width: int, height: int, data: Union[bytes, str], symbol: str)

Bases: object

Pythonic barcode object.

classmethod from_barcode(cls, barcode)

Initialize from pyzbar barcode object.

Parameters

barcode – A barcode (QR, CODE39, CODE128).

scale(self, factor: float)

Scale the barcode object by given factor.

The (top, left) is scaled accordingly.

Parameters

factor – Scaling factor for the barcode.

__str__(self)

Return str(self).

__repr__(self)

Return repr(self).

pypocquant.lib.barcode.detect(image: np.ndarray, expected_area=22000, expected_aspect_ratio=7.5, barcode_border=75, blur_size=(3, 3), morph_rect=(9, 3), mm_iter=1, qc=True, verbose=False)

Detect the barcode in the image.

Adapted from: https://www.pyimagesearch.com/2014/11/24/detecting-barcodes-images-python-opencv/

Returns the extracted barcode image, the coordinates of the extracted rectangle, the (possibly rotated) image, and (if qc is True) a copy of the (possibly rotated) image with the extracted rectangle coordinates overlaid on it.

Parameters

  • image (np.ndarray) – Image from which barcode should be read

  • expected_area (int) – Expected area for barcode.

  • expected_aspect_ratio (float) – Aspect ratio for barcode.

  • barcode_border (int) – Border of the barcode.

  • blur_size (tuple) – Kernel (3,3) by default for bluring the image.

  • morph_rect (tuple) – Kernel (9,3) by default for morph rect.

  • mm_iter (int) – Dilation & Eroding iterations.

  • qc (bool) – Bool, if true quality control images will be saved.

  • verbose (bool) – Bool, if true additional loggin info will be displayed.

Returns

barcode_img: The image of the barcode.

Returns

coordinates: The position and size coordinates of the barcode (x,y, w, h)/

Return type

coordinates: tuple

Returns

image: The image.

Returns

mask_image The mask of the image.

Return type

tuple

pypocquant.lib.barcode.rotate(image, angle)

Rotate the image by given angle in degrees.

Parameters

  • image – The image to be rotated.

  • angle – Rotation angle in degrees for the image.

Returns

image: Rotated image.

pypocquant.lib.barcode.calc_area_and_approx_aspect_ratio(contour)

Calculate area and approximate aspect ratio of a contour.

Parameters

contour – cv2.Contour.

Returns

area: Area of the contour.

Returns

aspect_ratio: Aspect ratio of the contour.

pypocquant.lib.barcode.rotate_90_if_needed(image)

Try to estimate the orientation of the image, and rotate if needed.

@TODO: This is not very robust so far.

Parameters

image – Image to be rotated by 90 degrees.

Returns

image: By 90 degrees rotated image.

pypocquant.lib.barcode.read_FID_from_barcode_image(image)

Read the FID string from the barcode image using pytesseract and decode the barcode itself using pyzbar.

Parameters

image – Image to read FID from barcode.

Returns

fid_tesseract: FID detected by tesseract (OCR).

Returns

fid_pyzbar: FID detected by pyzbar (barcode).

Returns

score: Score how well FID detection worked. For more details about the score read the manual.

pypocquant.lib.barcode.get_fid_from_barcode_data(barcode_data, barcode_type='CODE128')

Parse the output of pyzbar and retrieve the FID.

Parameters

  • barcode_data – Barcode data (zbar).

  • barcode_type – Type of barcode (CODE39, CODE128, QRCODE).

Returns

barcode: Decoded barcode as utf8.

pypocquant.lib.barcode.get_fid_from_box_image_using_ocr(box_img)

Use pytesseract to retrieve FID from the strip box image.

Parameters

box_img – Image of the QR code box.

Returns

fid_tesseract: FID detected by tesseract from image using OCR.

pypocquant.lib.barcode.try_extracting_barcode_from_box_with_rotations(box, scaling=(1.0, 0.5, 0.25), verbose=False, log_list=None)

Try extracting barcode from QR code box while scaling it for different orientations [0, 90, 180, -90].

Parameters

  • box – QR code box

  • scaling – Scaling factors.

  • verbose – Display additional logging information to the console.

  • log_list – Log list.

Returns

fid: FID number

Returns

log_list Appended Log list with current log information.

pypocquant.lib.barcode.try_extracting_barcode_with_rotation(image, angle_range=15, verbose=True, log_list: list = None)

Try extracting barcode from QR code box for a list of angles in the range of angle_range.

Parameters

  • image – Input image

  • angle_range (int) – Range of angles to rotate input images in degrees.

  • verbose – Display additional logging information to the console.

  • log_list – Log list.

Returns

fid: Extracted FID

Returns

angle: Rotation angle that led to FID detection

Returns

log_list: Appended log list.

pypocquant.lib.barcode.find_strip_box_from_barcode_data_fh(image, barcode_data, qr_code_border=30, qc=False)

Extract the box around the strip using the QR barcode data.

Parameters

  • image – Strip image.

  • barcode_data – Barcode data.

  • qr_code_border – Border around QR codes.

  • qc – Bool, if true quality control image will be saved.

Returns

box: Strip box.

Returns

qr_code_size: The size of the QR codes (qr_code_width, qr_code_height).

Returns

qc_image: Quality control image.

Returns

box_rect: Rectangle of the QR box.

pypocquant.lib.barcode.find_strip_box_from_barcode_data(image, barcode_data, qr_code_border=30, qr_code_spacer=40, barcode_border=80, qc=False)

Extract the box around the strip using the QR barcode data.

Parameters

  • image – Input image.

  • barcode_data – Barcode data

  • qr_code_border – Border around QR code on image.

  • qr_code_spacer – Spacer around QR code.

  • barcode_border – Border around barcode such as CODE128.

  • qc – Bool, if true quality control image will be saved.

Returns

box QR code box around strip

Returns

x_barcode Return the (x) coordinate of the left edge of the barcode rectangle.

Returns

qr_code_size: The size of the QR codes (qr_code_width, qr_code_height).

Returns

qc_image Quality control image.

pypocquant.lib.barcode.try_extracting_barcode_with_linear_stretch(image, lower_bound_range=(25), upper_bound_range=(98))

Try to extract the barcodes from the image by rescaling the intensity of the image with a linear stretch.

Parameters

  • image – Input image

  • lower_bound_range – Lower bound range.

  • lower_bound_range – tuple

  • upper_bound_range – Upper bound range.

  • upper_bound_range – tuple

Returns

””

Returns

gray

pypocquant.lib.barcode.try_getting_fid_from_code128_barcode(barcode_data)

Try finding a CODE 128 barcode in barcode data that should contain the patient FID.

Parameters

barcode_data – Barcode data

Returns

barcode: Decoded CODE128 barcode.

pypocquant.lib.barcode.try_get_fid_from_rgb(image)

Extract FID from rgb image.

Parameters

image – RGB image with FID.

Returns

fid: Detected FID as string.

pypocquant.lib.barcode.try_extracting_fid_and_all_barcodes_with_linear_stretch_fh(image, lower_bound_range=(0, 5, 15, 25, 35), upper_bound_range=(100, 98, 95, 92, 89), scaling=(1.0))

Try extracting the fid and all barcodes from the image by rescaling the intensity of the image with a linear stretch.

Parameters

  • image – Input image

  • lower_bound_range – Lower bound range.

  • lower_bound_range – tuple

  • upper_bound_range – Upper bound range.

  • upper_bound_range – tuple

  • scaling – Scaling factor

  • scaling – tuple

Returns

barcodes: Barcode object

Returns

fid: FID number

Returns

manufacturer: Manufacturer name.

Returns

plate: Plate info.

Returns

well: Well info.

Returns

user: Additional user data.

Returns

best_lb: Best lower bound.

Returns

best_ub: Best upper bound

Returns

best_score: Best score.

Returns

best_scaling_factor: Best scaling factor

Returns

fid_128: FID 128 code.

pypocquant.lib.barcode.try_extracting_all_barcodes_with_linear_stretch(image, lower_bound_range=(0, 5, 15, 25, 35), upper_bound_range=(100, 98, 95, 92, 89))

Try extracting the fid and all barcodes from the image by rescaling the intensity of the image with a linear stretch.

Parameters

  • image – Input image.

  • lower_bound_range – Lower bound range.

  • lower_bound_range – tuple

  • upper_bound_range – Upper bound range.

  • upper_bound_range – tuple

Returns

best_barcode_data

Returns

best_lb

Returns

best_ub

Returns

best_score

pypocquant.lib.barcode.rotate_if_needed_fh(image, barcode_data, image_log, verbose=True)

Rotate the image if the orientation is not the expected one.

Parameters

  • image – Input image.

  • barcode_data – Barcode data.

  • image_log – Image log list.

  • verbose (bool) – Bool, if true displays additional information to the console.

Returns

image_was_rotated: Bool, true if image was rotated.

Return type

image_was_rotated: bool

Returns

image: Rotated image.

Return type

tuple

pypocquant.lib.barcode.rotate_if_needed(image, barcode_data, image_log, verbose=True)

Rotate the image if the orientation is not the expected one.

Parameters

  • image – Input image.

  • barcode_data – Barcode data.

  • image_log – Image log list.

  • verbose (bool) – Bool, if true displays additional information to the console.

Returns

image_was_rotated: Bool, true if image was rotated.

Return type

image_was_rotated: bool

Returns

image: Rotated image.

Returns

image_log: Log for this image

Return type

tuple

pypocquant.lib.barcode.pick_FID_from_candidates(fid_pyzbar, fid_tesseract)

Selection of FID from candidates depending on if candidates contain a FID.

Parameters

  • fid_pyzbar – FID string determined with pyzbar.

  • fid_tesseract – FID string determined with tesseract.

Returns

fid FID number

Returns

score Score for the candidate determination.

pypocquant.lib.barcode.mask_strip(strip_gray, x_barcode, qr_code_extents)

Hide the barcode on the strip image.

Parameters

  • strip_gray – Image of the strip (POCT).

  • x_barcode – X coordinate of the barcode on the strip.

  • qr_code_extents – QR code extents on the strip.

Returns

strip_gray_masked Strip with QR code masked away.

Returns

background_value Background value used for strip masking.

pypocquant.lib.barcode.extract_strip_from_box(box, qr_code_width, qr_code_height, qr_code_spacer=40, slack=0)

Extract the strip from the strip box.

Parameters

  • box – Image of the QR code box.

  • qr_code_width – Width ot the QR code

  • qr_code_height – Height ot the QR code

  • qr_code_spacer – Horizontal and vertical distance between the internal edge of the QR codes and the beginning of the strip.

  • slack – Some buffer (subtracted from qr_code_spacer) to avoid cropping into the strip

Returns

strip Returns the extracted POCT strip as image matrix.

pypocquant.lib.barcode.get_fid_numeric_value_fh(fid)

Return the numeric value of the FID (as string).

A FID could be in the form ‘F0123456’. We want to preserve the leading 0 after we removed the ‘F’.

Parameters

fid (str) – FID number

Returns

fid:

pypocquant.lib.barcode.get_fid_numeric_value(fid)

Return the numeric value of the FID.

Parameters

fid (str) – FID number

Returns

filtered_fid: FID number as numeric

pypocquant.lib.barcode.get_box_rotation_angle(pt1, pt2, pt3)

Determine the the QR code box rotation angle

Parameters

  • pt1 – Coordinate corner 1

  • pt2 – Coordinate corner 2

  • pt3 – Coordinate corner 3

Returns

rot_angle Rotation angle in degree.

pypocquant.lib.barcode.align_box_with_image_border_fh(barcode_data, image)

Method to align QR code box with image border of the full image (old pipeline).

Parameters

  • barcode_data – QR code data

  • image – Image

Returns

image_rotated: Rotated image

Returns

angle Rotation angle in degrees.

pypocquant.lib.barcode.align_box_with_image_border(barcode_data, image)

Method to align QR code box with image border of the full image.

Parameters

  • barcode_data – QR code data

  • image – Image

Returns

image_rotated: Rotated image

Returns

angle Rotation angle in degrees.