In this paper， an active contour model combined with the hybrid signed pressure function is proposed to segment the images with intensity inhomogeneity or noise. Firstly， according to the position of the current active contour， we define a hybrid signed pressure function by using the global and local information of the image， which is a linear combination of a global pressure and a local one with adaptive weights. Then， an evolution equation of the active contour is constructed based on the hybrid signed pressure function. Finally， an alternating iteration algorithm is devolved to solve the model. Different synthetical， medical and real images are used to test the model. The experimental results show that the proposed model is robust to the initial contour and can effectively segment the images with intensity inhomogeneity or noises. Compared with other active contour models， the proposed model has the best performance.