The process of identifying chromosomes in metaphase images is demanding and it takes a lot of time. Thus, chromosome detection correctly in metaphase images is very important. It is proven that deep learning methods have been effective in object detection. Using a deep learning network, an automatic chromosome detection model is constructed for metaphase chromosome images. The model learns the chromosome images at the pixel level in less time. The detection results of the network verify the positive role of the model. This model of automatic chromosome detection based on deep learning methods may speed up the process of detecting anomalies.