How to Accurately Determine HSV Boundaries for Color Detection in OpenCV using cv2.inRange?

Determining HSV Boundaries for Color Detection with cv2.inRange in OpenCV

When using the cv2.inRange function in OpenCV for color detection, it is crucial to choose appropriate upper and lower HSV (Hue, Saturation, Value) boundaries to accurately identify the target color.

Problem Identification:

Consider the example of an image containing a coffee can with an orange lid, as shown below.

[Image of a coffee can with an orange lid]

The objective is to determine the HSV boundaries to isolate the orange lid. Initially, a range of (18, 40, 90) to (27, 255, 255) was attempted, but it yielded unexpected results.

Solution 1: Scale Conversion

It is important to note that different applications may use different scales for HSV values. OpenCV uses a scale of H: 0-179, S: 0-255, V: 0-255, while some other applications may use a scale of H: 0-360, S: 0-100, V: 0-100. Therefore, it is necessary to convert the HSV values accordingly.

Solution 2: Color Space Conversion

OpenCV uses the BGR (Blue, Green, Red) color format by default, while the image may be in RGB (Red, Green, Blue) format. To correctly convert the image to HSV, it is essential to use cv2.COLOR_BGR2HSV instead of cv2.COLOR_RGB2HSV.

Revised Code:

import cv2
import numpy as np

image = cv2.imread('kaffee.png')

# Revised HSV boundaries considering scale conversion
ORANGE_MIN = np.array([5, 50, 50], np.uint8)
ORANGE_MAX = np.array([15, 255, 255], np.uint8)

# Convert image to HSV color space
hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)

# Apply color filtering
mask = cv2.inRange(hsv, ORANGE_MIN, ORANGE_MAX)

# Save the masked image
cv2.imwrite('kaffee_out.png', mask)

This revised approach should provide more accurate results in isolating the orange lid of the coffee can.

The above is the detailed content of How to Accurately Determine HSV Boundaries for Color Detection in OpenCV using cv2.inRange?. For more information, please follow other related articles on the PHP Chinese website!