Thesis Details
Mobilní aplikace pro detekci graffiti tagů
Thesis focuses on the object recognition of images, using the principles of artificial intelligence. It solves the signature detection of authors in the field of art called graffiti. It concerns about basic problematic of this field, it also points to the use of computer vision followed by practical application on mobile devices, specifically on the Android platform. The selected neural network models was the ssdMobileNet_v2. The trained model achieves mAP accuracy of 73.5% meanwhile the IoU was set to 0.6. After the quantization process, the accuracy was reduced to 68.5%. The mobile application provides real-time detection and several other necessary functions for localization and data collection.
Computer vision, graffiti, mobile application, neural networks, TensorFlow, Android
Bidlo Michal, doc. Ing., Ph.D. (DCSY FIT BUT), člen
Čadík Martin, doc. Ing., Ph.D. (DCGM FIT BUT), člen
Křivka Zbyněk, Ing., Ph.D. (DIFS FIT BUT), člen
Rogalewicz Adam, doc. Mgr., Ph.D. (DITS FIT BUT), člen
@bachelorsthesis{FITBT19538, author = "P\v{r}emysl Chovane\v{c}ek", type = "Bachelor's thesis", title = "Mobiln\'{i} aplikace pro detekci graffiti tag\r{u}", school = "Brno University of Technology, Faculty of Information Technology", year = 2019, location = "Brno, CZ", language = "czech", url = "https://www.fit.vut.cz/study/thesis/19538/" }