An Improved ORB Feature Extraction Algorithm based on Enhanced Image and Truncated Adaptive Threshold

The ORB (Oriented fast and rotated brief, ORB) algorithm is limited by its inability to extract feature points or only a small number of when fixed threshold used in complex lighting conditions. To address this issue, study proposes method that combines image enhancement with truncated adaptive improve the extraction algorithm. Firstly, original converted grayscale. Secondly, enhanced applying Gaussian filtering for noise reduction, gamma brightness adjustment, unsharp masking operation. Then, segmented into subregions specified size, an improved OTUS employed calculate each subregion. Finally, are extracted using threshold. Experimental results show can significantly increase conditions, good accuracy, real-time performance, robustness.