Efficient graph-based algorithm is a new image segmentation method based on graph theory and it can segment an image at an extraordinary speed. However, it is easily influenced by the threshold value and the segmentation result is imprecise when dealing with the border and texture. Here, an improved algorithm was proposed, which had three main contributions: 1) RGB color space was replaced by Lab color space; 2) Laplacian operator was used to divide the edges of weighted graph into border edges and non-border edges, and those non-border edges were given priority; 3) the optimum threshold was evaluated based on uniformity measurement. The experimental results show that the improved algorithm is more accurate and adaptive than traditional graph-based algorithms, and the segmentation results are closer to that by human vision.%Graph-Based方法是基于图论的彩色图像分割算法中比较新颖的一种方法,且分割速度非常快.针对该算法对边缘和纹理处理效果不佳,且分割效果易受阈值影响的局限,改变了其颜色空间,结合拉普拉斯算子将带权图的边分为边缘边和非边缘边,优先处理非边缘边;再引入均匀性测度求取分割效果最佳的阈值.实验结果表明,相对于Graph-Based方法,改进的算法分割效果具有较好的准确性和适应性,更接近于人眼的感觉.
展开▼