The paper we are presenting is Offline Handwritten Character Recognition using Machine Learning. Because accessibility of a large amount of data as well as numerous algorithm upheavals has led to training machine learning effortlessly. The image segmentation is built on handwritten character recognition. We have used OpenCV for image processing, Tensorflow for training the Neural Network and used python programming language to develop this system. Machine Learning takes out concealed details that lie in the data. Machine Learning can be used to attain and foresee output for unrevealed data by which we are putting some mathematical functions and notions to reveal unrevealed details. One main application of Machine Learning is Pattern Recognition. Because of the large image data set, patterns are perceived by others. By utilizing these ideas, we have to a train computer to read alphabets and numeric characters in any language existing in an image dataset. There exist various methods using handwritten character recognition with which it can be recognized.