A new method combining microscope and microfluidic chip technology was proposed to realize the dynamic microscopic imaging of red blood cells, which was used to measure and analyze the multi-angle morphology of red blood cells (RBCs). Firstly, the microfluidic chip is used to make the red blood cells rotate or turn in a specific position by the contraction/expansion structure; then, using Matlab frame processing to collect the video image, through gray adjustment, median filtering, morphological processing, centroid method, object recognition, dynamic tracking, etc., to achieve the effective identification of blood cells and accurate classification and counting; finally, multi-angle morphological analysis was performed on the classification results, and the maximum diameter of RBCs with different morphology was measured. The accuracy of the proposed method is verified by analyzing the single cell multi-frame associated images. The blood of diabetic patients and normal people was tested by using the method of multi-angle morphological analysis of RBCs. The results showed that the average maximum diameter and the distribution width of diabetic erythrocytes were larger than that of normal erythrocytes. The method of microfluidic chip combined with image processing can realize high-throughput detection and multi-angle analysis of RBCs in diabetic patients and normal people, which provide a new method for multi-angle morphological measurement of other types of sample cells.