Correction of Real-Time Exposure in Hyperspectral Bands for Industrial Cameras

In this study, a new exposure correction method is presented for hyperspectral imaging. In the beginning, integration of the hardware parts to be used in the system is being done. Then a reference chart shows the minimum and maximum exposure values for each band is generated. Images with different exposure times have obtained in different hyperspectral bands using with an improved image acquisition interface. The various self-attributes that can represent the exposure situation have determined and a data set has created. The performance of Multilayer Sensor, Linear Regression and REPTree algorithms from artificial neural networks (ANN) models to determine the exposure quality is compared. The REPTree algorithm, which estimates the exposure situation 99.18%, is used for an application. It has been developed to determine the image with the highest exposure quality in the desired hyperspectral band and the results are discussed in this study.