sss3799/low-light-object-detection-

Low-Light-Object-Detection-

Abstract– This term paper presents the results of an investigation into the effect of varying exposure levels on the performance of an image classification model. The hypothesis was that training a model with a spectrum of black-and-white images with different exposure levels would improve its ability to classify images under low-light conditions. To test this hypothesis, we trained two models: one on a spectrum of images with varying exposure levels and another on low exposure images only. The results showed that the model trained on the spectrum of images achieved significantly better accuracy than the control model, indicating that exposure to diverse lighting conditions during training helped the model develop more robust representations of visual features. These findings have important implications for improving image classification algorithms in low-light environments and for understanding how the visual system adapts to changes in lighting conditions. The paper concludes with a discussion of the limitations of the study and suggestions for future research in this area.