Machine learning Exam.pdf
- 1. Detection of Fake currency note using Machine Learning Using Convolutional Neural Network Model is trained using different images of different currency notes such as 2000rs note, 500rs note. First of select a target size like how much the image should be like so that can scale it down to require more and fix number of bag sizes and define the classes but the classes is like which notes we have like 100rs, 500rs. After running this it trains the module then it can identify which is ever note we input like and which note it is can identified clearly. Block Diagram: False True Input Preprocessing True currency is detected Get features of currency from both the side Compare it using Machine learning algorithm Fake currency is detected Does features matches
- 2. Steps Implementation: Image Acquisition output comparison Pre- Processing Gray scale conversion Feature contraction
- 3. The steps used for image preprocessing are : Read image Resize image Remove noise(Denoise) Segmentation Morphology(smoothing edges) Step 1: Read image In this step, we save the path to our image dataset into a variable then created a function to load the folders containing images into arrays. Step 2: Resize image In this step in order to visualize the change, we are going to create two functions to display the images the first being a one to display one image and the second for two images. After that, we then create a function called processing that just receives the images as a parameter. Then some images may vary in size so we should give a base size for all images. Step 3: Remove noise To remove unwanted noise we use gaussian blur in this code. Gaussian function is used to blurring an image to reduce the image noise. Step 4: Segmentation In this step we are going to separate the background from foreground objects and further improve the segmenation with more noise removal. Step 6: Morphology In this step, we separate different objects in the image with markers.