Skip to content

Latest commit

 

History

History
57 lines (40 loc) · 2.6 KB

README.md

File metadata and controls

57 lines (40 loc) · 2.6 KB
Raw

Generalized Hough Transform Algorithm for Shape Detection (Algorithm 2)

Output using algo1

Output using algo1

--Output Using Algorithm 2 This algorithm detects shapes in images using the Generalized Hough Transform. It involves several steps including edge detection, gradient orientation computation, and gradient accumulation.

Key Components and Steps

1. Edge Detection and Gradient Orientation

  • Canny Edge Detection:
    • Uses the canny function to detect edges in the image, effective for identifying boundaries.
  • Gradient Orientation Calculation:
    • Computes gradient orientation using the Sobel filter, with gradients calculated in both x and y directions (dx and dy).

2. R-Table Construction

  • Building the R-Table:
    • Constructs an R-Table based on detected edges and their gradient orientations. It stores relative positions (vectors) of edge points with respect to a reference point.

3. Accumulation of Gradients

  • Accumulate Gradients Using R-Table:
    • Uses the R-Table to accumulate votes in an accumulator array. For each edge point, the function looks up corresponding vectors and increments the accumulator.

4. Overlay and Visualization

  • Overlay the Reference Image:
    • The overlay_reference_image function overlays the reference image on the query image at the detected position.
  • Visualization:
    • The test_general_hough function visualizes the results including the reference image, query image with detected points, the accumulator, and the final overlay.

5. Testing and Usage

  • Shape-to-Image Conversion:
    • Converts a set of shapes into binary images using the shapes_to_image function.
  • Testing:
    • The test function reads shape data from CSV files, converts them to images, and applies the Generalized Hough Transform.

Example Workflow

  1. Read and Convert Shape Data:

    • Read shape data from CSV files and convert them to binary images.

  2. Apply Generalized Hough Transform:

    • Apply the Generalized Hough Transform to the query image using the reference images.

  3. Visualize Results:

    • Visualize the results including the best match and the overlayed reference image.

Algo 2 Summary

  • Edge Detection: Uses Canny edge detection and Sobel filter.
  • R-Table Construction: Creates a lookup table for edge points.
  • Gradient Accumulation: Accumulates votes in an accumulator array.
  • Overlay and Visualization: Displays detected shapes and overlays the reference image..