Residual connection and multi-task multi-weight loss combined cell segmentation method and device

The invention discloses a residual connection and multi-task multi-weight loss combined cell segmentation method and device, and the method comprises the steps: (1) opening a medical cell image, intercepting cells with different styles and characteristics, and building a cell image data set; (3) making a cell label by using a label making program; (3) splitting the cell image data set into a training set, a test set and a verification set according to a proportion of 8: 1: 1; (4) using the cell data training set to train a multi-weight and residual error connection cell segmentation network; (5) using the test set to carry out loss verification evaluation on the training model of each round; (6) verifying the trained model by using a verification set, and finally evaluating the precision of the model; (7) judging a model evaluation result when the model evaluation precision is gt; executing the step (8) if the sample number is equal to 90% or the expectation is met, otherwise, executing the step (1) to increa