OpenCV Getting Started

1. Experiment Objective

Master OpenCV basics: image reading, display, saving, color space conversion, and other basic operations.

2. Experiment Principle

1. Keyword Explanation

2. Technical Architecture

This experiment is a pure offline Python script and does not depend on ROS2 or micro_ros_agent. The script runs in a Linux desktop environment, using the OpenCV library directly to read local images, perform pixel-level operations, and display the results through a GUI window.

Data flow: local image file → cv2.imread() loads it as a NumPy ndarray → OpenCV API processing → cv2.imshow() displays it → cv2.waitKey() waits for user interaction.

3. Experiment Steps

1. Hardware Preparation

2. Startup Method

This experiment provides three running modes, corresponding to different core operations:

Mode 1: Read and display an image

python3 ~/microros_ws/src/microros_v2_opencv_base_pkg/scripts/opencv_getting_started/opencv_read_show.py

Mode 2: Write/save an image

xxxxxxxxxx
python3 ~/microros_ws/src/microros_v2_opencv_base_pkg/scripts/opencv_getting_started/opencv_write.py

Mode 3: Pixel operations

xxxxxxxxxx
python3 ~/microros_ws/src/microros_v2_opencv_base_pkg/scripts/opencv_getting_started/opencv_pixel_manipulation.py

3. Operation Instructions

After running, a window pops up to display the image:

• opencv_read_show.py: reads and displays the original image, and prints size information to the terminal
• opencv_write.py: demonstrates image saving, writing the loaded image to a new file
• opencv_pixel_manipulation.py: modifies the pixel values of a specified row to black via NumPy array indexing

Press q to close all windows.

4. Stop Method

Press q to close the OpenCV window and exit the program.

4. Experiment Phenomena

• Successfully load and display the image
• Understand the usage of cv2.imread(), cv2.imshow(), and cv2.imwrite()

Original image: The test image used for all subsequent OpenCV operation demos, loaded and displayed via cv2.imread().

Result (copied image): Uses cv2.imwrite() to save the loaded image data as a new file, verifying the integrity of the image read/write pipeline.

Result (modified image): Directly modifies pixel values via NumPy array indexing (e.g., changing the color of an ROI region), demonstrating that OpenCV image data is essentially a NumPy ndarray.

5. Program Analysis

Source code path:

• Script directory: ~/microros_ws/src/microros_v2_opencv_base_pkg/scripts/opencv_getting_started/
• Test image: ~/microros_ws/src/microros_v2_opencv_base_pkg/scripts/images/yahboom.png

1. Core Logic Analysis

All three scripts follow the same execution flow: load image → process → display loop → release resources. The core code of each mode is described below one by one.

Image read and display (opencv_read_show.py)

​x
# Source file: ~/microros_ws/src/microros_v2_opencv_base_pkg/scripts/opencv_getting_started/opencv_read_show.py
import cv2
​
if __name__ == "__main__":
    # Read the image
    img = cv2.imread("~/microros_ws/src/microros_v2_opencv_base_pkg/scripts/images/yahboom.png", cv2.IMREAD_UNCHANGED)
    # Check whether the image was read successfully
    if img is None:
        print("Image read failed, please check that the images/yahboom.png path is correct")
        exit()
​
    print("Image size info:", img.shape)
​
    while True:
        # Display the image
        cv2.imshow("yahboom", img)
​
        # Wait for a key event, press 'q' to exit
        action = cv2.waitKey(10) & 0xFF
        if action == ord('q') or action == 113:
            break
​
    cv2.destroyAllWindows()

Key function notes:

• cv2.imread(path, flag): decodes an image file into a NumPy ndarray (BGR format); IMREAD_UNCHANGED preserves the original channels
• img.shape: returns the (height, width, channels) triple, used to get the image dimensions
• cv2.imshow(winname, img): creates and refreshes a GUI window to display the image
• cv2.waitKey(10): waits 10 ms and returns the key ASCII code; it is the core of the OpenCV GUI event loop
• cv2.destroyAllWindows(): closes all windows created by OpenCV

Pixel operations (opencv_pixel_manipulation.py)

xxxxxxxxxx
# Source file: ~/microros_ws/src/microros_v2_opencv_base_pkg/scripts/opencv_getting_started/opencv_pixel_manipulation.py
import cv2
​
if __name__ == '__main__':
    img = cv2.imread("~/microros_ws/src/microros_v2_opencv_base_pkg/scripts/images/yahboom.png", cv2.IMREAD_UNCHANGED)
    if img is None:
        print("Image read failed, please check that the images/yahboom.png path is correct")
        exit()
​
    (b, g, r) = img[100, 100]
    print(b, g, r)
​
    # Modify the pixel values of row 10 to black
    i = 10
    for j in range(1, 255):
        img[i, j] = (0, 0, 0)
​
    while True:
        cv2.imshow("yahboom", img)
        action = cv2.waitKey(10) & 0xFF
        if action == ord('q') or action == 113:
            break
​
    cv2.destroyAllWindows()

Key logic notes:

• img[y, x] returns the (B, G, R) triple; OpenCV uses BGR instead of RGB channel order
• img[i, j] = (0, 0, 0) directly assigns a value to modify a pixel; (0,0,0) is pure black
• The range(1, 255) loop sets the first 254 pixels of row 10 to black, forming a black horizontal line

2. Key Parameters

3. Node Notes

This experiment is a pure Python script and does not use ROS2 nodes. The script uses if __name__ == '__main__': as its entry point and runs independently in a desktop environment.

6. Common Problems