The head cabbage has three types according to its external ball shape, i.e., tip, flat and round shape types. The traditional identification method of cabbage ball shape is done artificially. A new method for rapid identification of cabbage ball shape was proposed using machine vision technology combined with BP neural network. Firstly, four absolute cabbage shape parameters were extracted, such as height, width, long axis and area, based on image processing technology. Five relative shape parameters were defined based on the above absolute parameters, which were ratio of height to width, circular degree, rectangle degree, ellipse degree and dome shape index. These nine parameters were used to describe the cabbage shape. Since the parameter ranges overlapped, the individual parameter did not have separating classification ability. Secondly, three recognition models of cabbage ball shape with BP neural network were established using three types of input datasets, four absolute parameters (long axis, height, width, area), five relative parameters (ratio of height to width, circular degree, rectangle degree, ellipse degree, dome shape index) and all above nine parameters. Each network had ten neurons in implicit layer, three neurons in output layer. Scaled conjugate gradient algorithm was used to train the network. The test results showed that the prediction accuracy of BP neural network model took four absolute parameters as the input was 62.5%, and the prediction accuracies of other two models were 100%. The model with relative parameters was relatively small and simple, and could shorten the time of network computing. Meanwhile, the center distance values of every two type training sample groups were computed, and the result showed that the model with all nine parameters had the biggest distance, which made the network be adapted to a wider sample spherical recognition.