In recent years, visual saliency detection has become a popular research topic. It can provide useful priorknowledge for high-level vision tasks, such as object detection and image classification. In this paper, agraph-based superpixel-wise similarity called ‘‘homology similarity” is proposed, which describes howlikely two superpixels belong to the same object or background region. A saliency detection model is thendeveloped based on the combination of homology distribution and improved color contrast. The homol-ogy distribution represents spatial compactness, while the color contrast characterizes color conspicuity.By combining these two saliency cues, the proposed model obtains more uniformly highlighted object-level saliency maps with fewer false positive noises. In the experiments, we evaluate our model and14 competing models (including traditional and state-of-the-art models) on the most popular datasetMSRA-1000 and 4 other publicly available datasets. Experimental results show that, compared with thesecompeting models, our model yields promising results.