摘要：Traffic accidents have a significant impact on dailylife, causing head injuries like skull fractures, brain damage,and so on. Many people fail to follow the safety regulations,such as riding a motorcycle without a helmet. The use ofmachine learning in brain haemorrhage research is extremelychallenging since it involves the collection of patient data fromcomputed tomography (CT) scan images. This study proposesa novel region-based segmentation approach for improving theaccuracy and efficiency of CT automated 3D image processingin the analysis of brain injuries. It is quite challenging to createa highly efficient superpixel method which maintains a strategicdistance from the segmentation and limited clusters of the pixelsin respect to the intensity boundaries. The approach reducescomputational costs, and the model achieves 97.79% accuracy insegmenting brain haemorrhage images. This study also guidesthe direction of future research in this domain.
关键词：SLIC algorithm; hybrid method; thresholding; region merging; segmentation
