dft_viewer.py

import cv2
import numpy as np


def compute_dft(gray: np.ndarray) -> np.ndarray:
    """Compute magnitude spectrum of grayscale image using DFT."""
    float_gray = np.float32(gray)
    dft = cv2.dft(float_gray, flags=cv2.DFT_COMPLEX_OUTPUT)
    dft_shift = np.fft.fftshift(dft)

    magnitude = cv2.magnitude(dft_shift[:, :, 0], dft_shift[:, :, 1])
    magnitude += 1e-8  # avoid log(0)
    magnitude_log = np.log(magnitude)
    magnitude_norm = cv2.normalize(magnitude_log, None, 0, 255, cv2.NORM_MINMAX)
    return magnitude_norm.astype(np.uint8)


def main():
    cap = cv2.VideoCapture(0)
    if not cap.isOpened():
        print("Error: Could not open webcam")
        return

    print("Webcam opened successfully!")
    print("Press 'q' to quit")

    while True:
        ret, frame = cap.read()
        if not ret:
            print("Error: Could not read frame")
            break

        gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
        spectrum = compute_dft(gray)

        # Convert to BGR to stack with original
        spectrum_bgr = cv2.cvtColor(spectrum, cv2.COLOR_GRAY2BGR)
        gray_bgr = cv2.cvtColor(gray, cv2.COLOR_GRAY2BGR)

        combined = np.hstack((frame, gray_bgr, spectrum_bgr))

        cv2.putText(combined, "Original", (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 255, 0), 2)
        cv2.putText(
            combined,
            "Grayscale",
            (frame.shape[1] + 10, 30),
            cv2.FONT_HERSHEY_SIMPLEX,
            0.7,
            (0, 255, 0),
            2,
        )
        cv2.putText(
            combined,
            "DFT Magnitude Spectrum",
            (frame.shape[1] * 2 + 10, 30),
            cv2.FONT_HERSHEY_SIMPLEX,
            0.7,
            (0, 255, 0),
            2,
        )

        cv2.imshow("Discrete Fourier Transform Viewer", combined)

        if cv2.waitKey(1) & 0xFF == ord("q"):
            break

    cap.release()
    cv2.destroyAllWindows()


if __name__ == "__main__":
    main()