karthi19sk/Fake-Currency-Detector

Ocatve code that performs morphological operations to identify if the given Ethiopian paper currency is real or fake based on a single security feature is implemented. # Fake-Currency-Detector Ocatve code that performs morphological operations to identify if the given Ethiopian paper currency is real or fake based on a single security feature is implemented. Security feature: *Every Original Ethiopian currency has a vertical strip in the centre that contain exactly a set of pixels forming one connected component. Step by step procedure: 1. The input image is compressed to 320*120 pixels or closer resolution. 2. The image is cropped to obtain pixels bw 103-108 horizontally, containing the vertical strip. 3. The noise due to dirt and flashlight are removed by selecting the pixels above a fixed saturation thrseshold and below a fixed intensity threshold respectively. 4. imclose(): erosion followed by dilation removes additional noises to give a sharp edged image. 5. bwareaopen(img,15): removes all disconnected components or smaller connected components that contain less than 15 pixels. 6. bwlabel(areaopen): Returns the number of connected pixels containing 15 or more pixels. 7. real/fake: if there exists 1 connected set, then the note is real else it is fake.