Canine Disease Detection and Classification Using Texture-Based Feature Extraction

All parties engaged in the transmission of an infectious illness must be aware of it in order to study and control it. In the absence of cross-contamination from other domestic or wild animals, a reservoir, also known as a maintenance host, "is able to sustain an infection in a specific region." One of the worst illnesses seen in people across the world is canine sickness. The number of instances is rising every year, and so are the many technologies, methods, and approaches for diagnosis. To circumvent these problems, our study creates a multi-directional image pattern representation for the accurate detection of neighbouring pixels. This might be accomplished using the Intensity Distributional Texture Pattern (IDTP) based feature extraction method. To identify the crucial pixels between the IDTP's centre and its neighbours, they were split into blocks. The cell's weight values were then calculated using that magnitude. These are used across the whole image matrix. This was then used to build up the output matrix, which represents the image pattern. The histogram of the pattern was then used to construct the feature vector of the image. This was then classified by using a neural network model to predict the class label of different canine coccidiosis sickness types.