Image Binarization - Color Image

Example Results

Run the example code in this section [Source Code/06.cv_lite/14.rgb888_threshold_binary.py]

In this section, we will use the cv_lite extension module to implement RGB888 format binarization on an embedded device.

Code Overview

Importing Modules

import time, os, sys, gc
from machine import Pin
from media.sensor import * # Camera interface
from media.display import * # Display interface
from media.media import * # Media manager
import _thread
import cv_lite # cv_lite extension with binary threshold
import ulab.numpy as np

Setting the Image Size

xxxxxxxxxx
image_shape = [480, 640] # Height x Width

Define the image resolution to 480x640 (Height x Width). Width). The camera and display module will be initialized based on this size later.

Initialize the camera (RGB888 format)

xxxxxxxxxx
sensor = Sensor(id=2, width=640, height=480)
sensor.reset()
sensor.set_framesize(width=640, height=480)
sensor.set_pixformat(Sensor.RGB888)

• Initialize the camera, setting the resolution to 640x480 and the frame rate to an unspecified default.
• Resize the output frame to 640x480 and set it to RGB888 format (a three-channel color image, suitable for binarization of color images).

Initialize the display module

xxxxxxxxxx
Display.init(Display.ST7701, width=640, height=480, to_ide=True, quality=50)

Initialize the display module, using the ST7701 driver chip, with a resolution consistent with the image size. to_ide=True indicates that the image will be simultaneously transmitted to the IDE for virtual display, and quality=50 sets the image transmission quality.

Initialize the media manager and start the camera

xxxxxxxxxx
MediaManager.init()
sensor.run()

Initialize the media resource manager and start the camera to capture the image stream.

Set binarization parameters

xxxxxxxxxx
thresh = 130 # Threshold value
maxval = 255 # Maximum value (white)
clock = time.clock() # Start FPS timer

• thresh: Binarization threshold. Pixels below this value will be set to 0 (black), while pixels above or equal to this value will be set to maxval (white).
• maxval: Maximum pixel value after binarization, typically set to 255 to represent white.
• clock: Used to calculate frame rate.

Image processing and binarization

​x
while True:
    clock.tick()
​
    # Capture a frame
    img = sensor.snapshot()
    img_np = img.to_numpy_ref()
​
    # Call binary threshold function (returns ndarray)
    binary_np = cv_lite.rgb888_threshold_binary(image_shape, img_np, thresh, maxval)
​
    # Construct grayscale image for display
    img_out = image.Image(image_shape[1], image_shape[0], image.GRAYSCALE,
    alloc=image.ALLOC_REF, data=binary_np)
​
    # Show image
    Display.show_image(img_out)
​
    # Garbage collection & print FPS and print FPS
    gc.collect()
    print("binary:", clock.fps())

• Image Capture: Acquire an image frame using sensor.snapshot() and convert it to a NumPy array reference using to_numpy_ref().
• Binarization: Call cv_lite.rgb888_threshold_binary() to perform binarization, returning a NumPy array of the binarized grayscale image.
• Image Packaging and Display: Pack the binarized result into a grayscale image object and display it on the screen or in an IDE virtual window.
• Memory Management and Frame Rate Output: Call gc.collect() to clear memory and print the current frame rate using clock.fps().

Resource Release

xxxxxxxxxx
sensor.stop()
Display.deinit()
os.exitpoint(os.EXITPOINT_ENABLE_SLEEP)
time.sleep_ms(100)
MediaManager.deinit()

Parameter Adjustment Instructions

• thresh: Threshold. A higher value requires brighter pixels to turn white. Adjust based on image brightness. It is recommended to test from 100 to 200 to filter noise or highlight the target.
• maxval: Maximum value. Usually kept at 255. If you need to customize the output intensity, you can adjust it, but generally it does not need to be changed.