Image processing features extraction on fish behaviour

This chapter demonstrates the pipeline fromdata collection until classifier 2 models that achieve the best possible model in identifying the disparity between 3 hunger states. The pre-processing segment describes the features of the data sets 4 obtained by means of image processing. The method includes the simple moving 5 average (SMA), downsizing factors, dynamic timewarping (DTW) and clustering by 6 the k-meansmethod. This is to rationally assign the necessary significant information 7 from the data collected and processed the images captured for demand feeder and 8 fish motion as a synthesis for anticipating the state of fish starvation. The selection of 9 features in this study takes place via the boxplot analysis and the principal component 10 analysis (PCA) on dimensionality reduction. Finally, the validation of the hunger 11 state will be addressed by comparing machine learning (ML) classifiers, namely the 12 discriminant analysis (DA), support vector machine (SVM) and k-nearest neighbour 13 (k-NN). The outcome in this chapter will validate the features fromimage processing 14 as a tool for identifying the behavioural changes of the fish in school size