In this work, an automatic system is created, which consists of three parts, namely, pre-processing, segmentation, and classification. The pre-processing part consists of changing color space, noise reduction, removal of unwanted objects: this uses grey conversion, morphological operations, and filtering. The segmentation part consists of separating red blood cells and saving them to other locations: this part consists of applying various techniques such as Otsu's thresholding and distance transform. Finally, in the classification part, cells are identified whether they are healthy or parasitic. A pre-trained convolution neural network model VGG16 is used for the classification task.