Skip to content

Card Patches Detection

Introduction

This package has built-in support for detecting card patches in an image. This feature is useful for detecting color checker cards in an image. The detection is based on the YOLOv8 model. You can use the YOLOv8CardDetector or MCCardDetector class to detect card patches in an image.

| | | | |

If you don't have image to test

You can download the sample image from the following link: 

curl -L -o input_image.jpg "https://drive.google.com/uc?export=download&id=1syOqw9kC0tt01p7yEobU4MeLfh336DZA"

Usage YOLOv8CardDetector

import cv2

from color_correction import YOLOv8CardDetector
from color_correction.schemas.det_yv8 import DetectionResult

image = cv2.imread("your_image_path")
detector = YOLOv8CardDetector(
    conf_th=0.15,
    iou_th=0.7,
    use_gpu=False, # (1)
)

result: DetectionResult = detector.detect(image=image)
drawed_image = result.draw_detections(image=image)
cv2.imwrite("drawed_detection.jpg", drawed_image)
  1. 💬 The model runs using onnx , which supports both CPU and GPU. The model will be automatically downloaded if not already present on your system.
{
    "boxes": [
        (366, 426, 441, 500), // (1)
        ...,
        (248, 138, 805, 545),
    ],
    "segments": null,
    "scores": [
        0.9755859375,
        ...,
        0.8583984375,
    ],
    "class_ids": [
        0, // (2)
        ...,
        1, // (3)
    ],
}
  1. 💬 (x1, y1, x2, y2): coordinates of the bounding box top-left and bottom-right corners
  2. 💬 for patch object
  3. 💬 for card object

Detection result

The model detects two types of objects: a complete card and individual color patches (24 patches per card). Click the image to view full resolution detection results.

Usage MCCardDetector

import cv2

from color_correction import MCCardDetector
from color_correction.schemas.det_yv8 import DetectionResult

image = cv2.imread("your_image_path")
detector = MCCardDetector(conf_th=0.15)

result: DetectionResult = detector.detect(image=image)
drawed_image = result.draw_detections(image=image)
cv2.imwrite("drawed_detection.jpg", drawed_image)
  1. 💬 The model runs using onnx , which supports both CPU and GPU. The model will be automatically downloaded if not already present on your system.
{
    "boxes": null,
    "segment": [
        [
            [0.0, 0.0],
            [0.0, 1.0],
            [1.0, 1.0],
            [1.0, 0.0],
        ],
        ...
        [
            [0.0, 0.0],
            [0.0, 1.0],
            [1.0, 1.0],
            [1.0, 0.0],
        ],
    ], // (1)
    "scores": [
        0.9755859375,
        ...,
        0.8583984375,
    ],
    "class_ids": [
        0, // (2)
        ...,
        1, // (3)
    ],
}
  1. 💬 [x1, y1, x2, y2]: coordinates segment
  2. 💬 for patch object
  3. 💬 for card object

Detection result

The model detects two types of objects: a complete card and individual color patches (24 patches per card). Click the image to view full resolution detection results.

Reference