LLTO: Towards efficient lesion localization based on template occlusion strategy in intelligent diagnosis

•A lesion localization method based on template occlusion (LLTO) in disease images is proposed.•The candidate boxes representation model and disease discrimination model are trained by OpenCV and Tensorflow.•The candidate boxes representation model are designed to generate the lesion candidate boxes.•The disease discrimination model is used to select the true lesion areas from candidate boxes. Lesion location information helps physicians have a better understanding of a disease, assists them in diagnosis and therapy, and increases the likelihood of a disease being cured. In this paper, we propose an efficient lesion localization method based on template occlusion (LLTO) for locating lesion areas in disease images. First, the OpenCV cascade classifier is used to train the candidate boxes representation model and TensorFlow is used to train the disease discrimination model. Second, to implement the lesion location task, the lesion candidate boxes are generated by the candidate boxes representation model in the test image. Third, the disease discrimination model is used to select the true lesion areas from candidate boxes and integrate them to obtain the image marked with lesion locations. The experiment proves the efficiency and effectiveness of our method.