Counting in Intracellular Images Using Partial Least Squares Regression and Correlation between Features
Shohei Kumagai, Kazuhiro Hotta · 2013
Light spot counting in intracellular images is important for investigating the cause diseases. However, light spots are manually counted by human observers now. This paper proposes an automatic light spot counting method by computer. To count light spots, we use partial least squares regression and correlation between 2 different types of features using higher-order local autocorrelation. The proposed method gives higher accuracy than counting by support vector regression and ImageJ. The effectiveness of our method is shown by experiments.