TY - JOUR N2 - This paper presents a deep learning-based image texture recognition system. The methodology taken in this solution is formed in a bottom-up manner. It means we swipe a moving window through the image in order to categorize if a given region belongs to one of the classes seen in the training process. This categorization is done based on the Deep Neural Network (DNN) of fixed architecture. The training process is fully automated regarding the training data preparation, investigation of the best training algorithm, and its hyper-parameters. The only human input to the system is the definition of the categories for further recognition and generation of the samples (region markings) in the external application chosen by the user. The system is tested on road surface images where its task is to categorize image regions to a different road category (e.g. curb, road surface damage, etc.) and is featured with 90% and above accuracy. L1 - http://www.czasopisma.pan.pl/Content/118373/PDF/25_D1503-1511_01686_Bpast.No.68-6_29.12.20_OK.pdf L2 - http://www.czasopisma.pan.pl/Content/118373 PY - 2020 IS - No. 6 EP - 1511 DO - 10.24425/bpasts.2020.135395 KW - deep learning KW - texture segmentation KW - artificial intelligence A1 - Kapela, R. VL - 68 DA - 31.12.2020 T1 - Texture recognition system based on the Deep Neural Network SP - 1503 UR - http://www.czasopisma.pan.pl/dlibra/publication/edition/118373 T2 - Bulletin of the Polish Academy of Sciences Technical Sciences ER -