Diverse Image Processing Techniques for Machine Learning Methods

Machine learning is a trending topic in the area of computer vision, which makes the machine able to learn about it without being expressly programmed using various algorithms. When a model is designed to predict future-based on historical data using machine learning algorithms, data is vital because high quality data is needed to run the model efficiently and to predict highly accurate results. Machine learning algorithms receive data in many forms, like audio (speech recognition), images (feature recognition), numeric data (like height and weight) etc. When it comes to computer vision, the machine learning algorithms must be able to understand the image data, and before the images are processed to the algorithms it should be pre-processed (cleaned and set to the desired format) using various image pre-processing techniques. This research paper presents various image processing techniques, literature reviews on various image processing techniques and image PSNR value after applying median filter, grayscale filter and box filters.