Join Our 5-Week ML/AI Engineer Interview Bootcamp 🚀 led by ML Tech Leads at FAANGs
147. Color space conversion
GeneralImplement an RGB → Grayscale color space conversion, a common preprocessing step in computer vision. Given an image of RGB pixels, convert it to a 2D grayscale image using a weighted luminance formula.
The grayscale intensity is defined as:
Requirements
Implement the function
Rules:
- Convert each pixel
[R, G, B]into a single grayscale value using the formula above. - Return a 2D NumPy array.
- Do not use any prebuilt image/color conversion utilities (e.g., OpenCV, PIL).
- Keep the output values as floats (no rounding required).
Example
Output:
| Argument | Type |
|---|---|
| image | np.ndarray |
| Return Name | Type |
|---|---|
| value | np.ndarray |
Constraints
Input array shape must be (H, W, 3).
Use NumPy vectorization; no OpenCV/PIL utilities.
Return NumPy array.
Hint 1
Ensure input image is a NumPy array of floats with shape (H, W, 3).
Hint 2
Use the luminance weights w = np.array([0.299, 0.587, 0.114]) and combine them with the RGB channels by taking a weighted sum across the last axis (e.g., image @ w or np.tensordot).
Hint 3
The result should be a NumPy array of shape (H, W).
Roles
Companies
GeneralLevels
Tags
Python Editor