askvity

How do I reduce the contrast of an image in Matlab?

Published in Image Processing 5 mins read

To reduce the contrast of an image in Matlab, you can adjust the intensity values of the pixels closer together, effectively reducing the difference between the darkest and brightest parts of the image. Here are a couple of common methods:

1. Using imadjust Function

The imadjust function is a straightforward way to adjust the contrast of an image. You can specify the input and output intensity ranges.

% Read the image
img = imread('your_image.jpg');

% Adjust the contrast
% syntax: imadjust(image,[low_in high_in],[low_out high_out],gamma)
% To reduce contrast, map a smaller input range to a larger output range

% Example: Reduce contrast by mapping a smaller input range to the full output range
img_low_contrast = imadjust(img, [0.3 0.7], [0 0.1]); % maps input intensities between 0.3 and 0.7 to output intensities between 0 and 0.1

% Display the original and adjusted images
figure;
subplot(1,2,1); imshow(img); title('Original Image');
subplot(1,2,2); imshow(img_low_contrast); title('Low Contrast Image');

Explanation:

  • imread('your_image.jpg'): Reads your image file. Replace 'your_image.jpg' with the actual filename.
  • imadjust(img, [low_in high_in], [low_out high_out], gamma): This function maps the intensity values in the input image img to new values.
    • [low_in high_in] specifies the range of intensity values in the original image that you want to adjust. Intensity values below low_in map to low_out, and intensity values above high_in map to high_out.
    • [low_out high_out] specifies the desired range of intensity values in the output image.
    • gamma (optional, defaults to 1) specifies the shape of the mapping curve. gamma < 1 makes the image brighter, while gamma > 1 makes the image darker.
  • To reduce contrast, map a smaller range of input intensities to a larger range of output intensities, or decrease the overall range of output intensities.

2. Using Histogram Equalization (with Limitations)

Histogram equalization typically increases contrast, but it can be adapted to reduce contrast if applied carefully. The basic approach is to limit the amount of equalization.

% Read the image
img = imread('your_image.jpg');

% Apply histogram equalization with clipping (to limit contrast enhancement)
% Convert to grayscale if necessary
if size(img, 3) == 3
    img_gray = rgb2gray(img);
else
    img_gray = img;
end

img_low_contrast = histeq(img_gray); %histogram equalization

% Display the original and adjusted images
figure;
subplot(1,2,1); imshow(img_gray); title('Original Image');
subplot(1,2,2); imshow(img_low_contrast); title('Low Contrast Image');

Explanation:

  • histeq(img_gray): Performs histogram equalization, which generally increases contrast by redistributing pixel intensities to utilize the full dynamic range. Because equalization tends to increase contrast, this method alone may not be ideal for reducing contrast.

Important Note on Histogram Equalization: The provided code using simple histogram equalization will generally increase contrast. To effectively reduce contrast using histogram-based techniques, consider:

  • Limiting equalization: Use techniques that constrain how much the histogram is modified. This involves custom implementations rather than directly using histeq.
  • Contrast Limited Adaptive Histogram Equalization (CLAHE): While CLAHE is generally for contrast enhancement, its parameters can be tuned to slightly reduce local contrast in some cases, though it is primarily used to avoid over-enhancement.

3. Directly Manipulating Pixel Values

You can directly scale and shift the pixel values to reduce contrast.

% Read the image
img = imread('your_image.jpg');

% Convert to double for calculations
img_double = im2double(img);

% Reduce contrast by scaling the pixel values closer to the mean intensity
mean_intensity = mean(img_double(:));
contrast_factor = 0.5;  % Adjust this factor (0 < factor < 1) to control contrast reduction

img_low_contrast = contrast_factor * (img_double - mean_intensity) + mean_intensity;

% Clip values to be within [0, 1]
img_low_contrast = max(0, min(1, img_low_contrast));

% Convert back to uint8 for display (if needed)
img_low_contrast = im2uint8(img_low_contrast);


% Display the original and adjusted images
figure;
subplot(1,2,1); imshow(img); title('Original Image');
subplot(1,2,2); imshow(img_low_contrast); title('Low Contrast Image');

Explanation:

  • im2double(img): Converts the image to double precision, allowing for floating-point arithmetic.
  • mean(img_double(:)): Calculates the mean intensity of all pixels in the image.
  • contrast_factor * (img_double - mean_intensity) + mean_intensity: Scales the difference between each pixel's intensity and the mean intensity by contrast_factor. A contrast_factor between 0 and 1 reduces the contrast. This scaling pushes pixel values closer to the mean.
  • max(0, min(1, img_low_contrast)): Clips the pixel values to the range [0, 1], ensuring they are valid for display.
  • im2uint8(img_low_contrast): Converts the image back to the uint8 data type, which is commonly used for displaying images.

The most effective way to reduce contrast depends on the specific image and the desired outcome. Try adjusting the parameters in each method to achieve the desired result.

Related Articles