Malaria Detection Using Image Processing And Machine Learning

Malaria which has now become a common human disease is diagnosed in the present scenario starting with a clinical screening and then by medical treatment. Automated classification of malaria parasites using images is a bit challenging task due to the large amount of variability found in the display of skin abrasion. Many deep convolution neural networks show possibility for general highly variable tasks across many fine-grained object categories. Here we are working by using CNN network and datasets. CNN which is known as convolutional neural network is used to differentiate images on the basis of image pixel patterns. Our model mostly focusses on image processing using keras image generator for creating real time images. We basically train our model to easily differentiate between positive and negative images from the given datasets. Detecting malaria by image processing is very fast and reliable method as it does not require any experience. The main aim of using image processing is that our model can detect cells from multiple images taken from microscope via thin blood smear and detect them as positive and negative human blood cell and also it performs classification on human blood cell by using deep learning.