CH-4 – Features Extraction – Part-4

Important Note: This article is part of the series in which TechReport.us discusses the theory of Video Stream Matching.

In the realm of Video Stream Matching, four essential features are integral:

1. Histogram
2. Edge
3. Slope
4. Wavelet

Here, we will delve into the algorithms associated with each of these features one by one.

4.4.1 Histogram Feature Extraction

4.4.1.1 Algorithm 2

As outlined in Chapter 3.

Input:

• BMP Image

Output:

• Array of numerical values defining the Histogram

Working:

1. The algorithm takes an image as input.
2. If the image is not already in grayscale, it is converted into grayscale.
3. The image is then divided into tiles, and the histogram of each tile is calculated.
4. The calculated histograms are stored in columns and rows, forming a 2D array that represents the result.

Algorithm Steps:

• Step 0: Start
• Step 1: Take input of an image (im).
• Step 2: Check if (im) is in grayscale. If not, convert (im) into grayscale.
• Steps 3-8: Loop through tiles of the image, extract sub-images, calculate histograms (H), increment the number of bins (NumBins), and store the histograms in the resultant array (ResArray).
• Step 9: Return the resulting array (ResArray).
• Step 10: Stop