Add contrast to area detector streaming

[SeijDeLeon]

Images streamed from non xray detector cameras (used to view samples or other views in the hutch) can come across very dark and difficult to see.

ImageJ does some auto scaling or contrast adjustment so that the largest intensities are scaled up.

We could perform this auto scaling on the python backend before it gets to the browser, or also provide some contrast scaling directly in the dom canvas element. There are some algorithms done by ALS Computing specifically to improve contrast for detector images, which may be useful here or may take too long to process live feeds.

[SeijDeLeon]

From als-computing:

# image_processing.py
import os
import cv2
import numpy as np

def equalize_16bit_histogram(img):
    """Performs histogram equalization for a 16-bit grayscale image."""
    hist, bins = np.histogram(img.flatten(), 65536, [0, 65536])
    cdf = hist.cumsum()
    cdf_normalized = (cdf - cdf.min()) * 65535 / (cdf.max() - cdf.min())
    cdf_normalized = cdf_normalized.astype('uint16')

    img_equalized = np.interp(img.flatten(), bins[:-1], cdf_normalized)
    return img_equalized.reshape(img.shape).astype(np.uint16)

def convert_to_8bit(img):
    """Converts a 16-bit image to 8-bit by normalizing values."""
    img_8bit = (img / 256).astype('uint8')  # Scale from 16-bit to 8-bit
    return img_8bit

def process_image(args):
    """Loads, processes, and saves a single image."""
    input_path, output_path = args

    img = cv2.imread(input_path, cv2.IMREAD_UNCHANGED)
    if img is None:
        print(f"❌ Failed to load: {input_path}")
        return

   # if img.dtype != np.uint16:
    #    print(f"⚠️ Skipping (Not 16-bit): {input_path}")
     #   return

    print(f"✅ Processing: {input_path}")

    # Apply histogram equalization
    img_eq_16bit = equalize_16bit_histogram(img)

    # Convert to 8-bit
    img_eq_8bit = convert_to_8bit(img_eq_16bit)

    # Ensure the output directory exists
    os.makedirs(os.path.dirname(output_path), exist_ok=True)

    # Save the processed 8-bit image
    cv2.imwrite(output_path, img_eq_8bit)
    print(f"✅ Saved: {output_path}")

sample:

process_image((input_path, output_path))