changelog.d/20241121_062608_neildencklau_opencv_image_clahe.md (1)

1-4: Enhance the changelog entry with more details.

While the entry correctly documents the addition of CLAHE, it could be more informative for users by including:

• The purpose (improved local contrast enhancement)
• Key configurable parameters (clip limit, tile grid size)
• Reference to the issue being addressed ([Feature request] Using CLAHE for Histogram equalization #4929)

Consider expanding the entry like this:

 ### Added
 
-Added contrast limited adaptive histogram equalization tool
-(<https://github.com/cvat-ai/cvat/pull/8730>)
+- Added contrast limited adaptive histogram equalization (CLAHE) tool for improved local contrast enhancement
+  - Configurable parameters: clip limit, tile rows, and tile columns
+  - Better visibility for images with varying lighting conditions
+  - Addresses image contrast issues (#4929)
+  (<https://github.com/cvat-ai/cvat/pull/8730>)

cvat-ui/src/utils/image-processing.tsx (1)

Line range hint 1-45: Add test coverage for the new CLAHE filter.

While the enum addition is straightforward, we should ensure proper test coverage for the CLAHE functionality, especially around the hasFilter function with the new filter alias.

Would you like me to help generate test cases for:

1. Verifying hasFilter works with the new CLAHE alias
2. Testing filter registration/configuration
3. Integration tests with the OpenCV control component

cvat-ui/src/utils/opencv-wrapper/opencv-wrapper.ts (2)

304-304: Consider adding error handling for OpenCV initialization

While the implementation is correct, consider adding error handling for cases where OpenCV initialization might fail. This would be consistent with the error handling pattern used in other methods of the class.

 clahe: () => new CLAHEImplementation(this.cv),
+clahe: () => {
+    try {
+        return new CLAHEImplementation(this.cv);
+    } catch (error) {
+        console.error('Failed to initialize CLAHE:', error);
+        throw new Error('Failed to initialize CLAHE implementation');
+    }
+},

---

Line range hint 1-324: Add unit tests for CLAHE functionality

The implementation looks solid, but there are no tests covering the new CLAHE functionality. Consider adding unit tests to verify:

1. Proper initialization of CLAHE
2. Error handling scenarios
3. Integration with the OpenCV wrapper
4. Parameter validation

This aligns with the PR objectives which mention that tests haven't been added yet.

Would you like me to help generate unit test cases for the CLAHE implementation?

site/content/en/docs/manual/advanced/ai-tools.md (2)

Line range hint 286-312: Consider adding context about limitations

The section effectively explains histogram equalization, but it would be helpful to briefly mention its limitations that led to implementing CLAHE. This context would help users understand when to use each method.

Add a note about limitations, for example:

It is useful in images with backgrounds
and foregrounds that are bright or dark.

+ Note: While histogram equalization works well for many images, it applies the enhancement globally.
+ This may not be optimal for images with varying lighting conditions across different regions.
+ In such cases, consider using CLAHE (see below).

To improve the contrast of the image, do the following:

---

314-314: Fix typo in CLAHE description

There's a spelling error in the text.

-Contrast Limited Adaptive Histogram Equalization (CLAHE) increases contrast by applying clipped histogram equalization to multiple tiles across the input image. In images where there are both very bright and very dark regions, this improves contrast in the dark regions without loosing contrast in the bright ones.
+Contrast Limited Adaptive Histogram Equalization (CLAHE) increases contrast by applying clipped histogram equalization to multiple tiles across the input image. In images where there are both very bright and very dark regions, this improves contrast in the dark regions without losing contrast in the bright ones.

cvat-ui/src/components/annotation-page/standard-workspace/controls-side-bar/opencv-control.tsx (4)

88-90: Consider using constants for CLAHE default values

The default values for CLAHE parameters could be defined as class constants or exported constants, making them easier to maintain and reuse.

+ const DEFAULT_CLAHE_CLIP_LIMIT = 40;
+ const DEFAULT_CLAHE_TILE_SIZE = 8;

class OpenCVControlComponent extends React.PureComponent<Props & DispatchToProps, State> {
    // ...
    this.state = {
-     claheClipLimit: 40,
-     claheTileColumns: 8,
-     claheTileRows: 8,
+     claheClipLimit: DEFAULT_CLAHE_CLIP_LIMIT,
+     claheTileColumns: DEFAULT_CLAHE_TILE_SIZE,
+     claheTileRows: DEFAULT_CLAHE_TILE_SIZE,
    };

Also applies to: 163-165

---

620-621: Consider adding more descriptive tooltip text

The current tooltip text is technical. Consider adding a brief explanation of what CLAHE does for users who might not be familiar with the technique.

- title='Contrast Limited Adaptive Histogram Equalization'
+ title='Enhance local contrast (CLAHE) - Improves image details in both bright and dark regions'

---

655-675: Enhance input validation and user feedback

The numeric inputs could benefit from:

1. Tooltips explaining the impact of each parameter
2. Immediate visual feedback when values are out of optimal ranges
3. Debounced updates to prevent excessive re-rendering

Consider wrapping the InputNumber components in a custom component that handles these improvements:

interface CLAHEInputProps {
  label: string;
  value: number;
  min: number;
  max: number;
  onChange: (value: number) => void;
  tooltip: string;
  defaultValue: number;
}

const CLAHEInput: React.FC<CLAHEInputProps> = ({
  label,
  value,
  min,
  max,
  onChange,
  tooltip,
  defaultValue
}) => {
  const debouncedOnChange = useMemo(
    () => debounce(onChange, 300),
    [onChange]
  );

  return (
    <Tooltip title={tooltip}>
      <InputNumber
        value={value}
        min={min}
        max={max}
        onChange={(v) => debouncedOnChange(v || defaultValue)}
        status={value > max * 0.8 ? 'warning' : undefined}
      />
    </Tooltip>
  );
};

Also applies to: 681-701, 706-726

---

619-730: Add tests for CLAHE functionality

The new CLAHE feature would benefit from unit tests covering:

1. Parameter validation
2. UI interaction
3. Integration with OpenCV wrapper
4. Error handling

Would you like me to help create a test suite for the CLAHE implementation?

cvat-ui/src/utils/opencv-wrapper/clahe.ts (1)

70-70: Consider rethrowing the original error to preserve stack trace

Throwing a new error with e.toString() loses the original stack trace, which can hinder debugging. Rethrow the original error instead.

Apply this change:

- throw new Error(e.toString());
+ throw e;

Configuration used: CodeRabbit UI
Review profile: CHILL