Thesis Details
Aplikace s rozpoznáváním jídla
The bachelor thesis deals with the design and analysis of a forward deep neural network with a convolutional core, to improve the food recognition performance by a standing learning algorithm with a teacher, which adaptively adjusts its own constraint weights based on gradient optimization results in the propagation of the loss function to converge to ideal spatial solution. The main benefit of this work is a new modular platform running on a polyglot compiler, which is suitable for supporting and extending a proactive module with indefinite web application functionality, where all data-providing services are built on advances in virtual computing studies. The goal representation method was to create a solution that provides a responsively generated user interface for web browsers for which stably independent servers were prepared by parallel system research.
LaTeX, Vim, PGP, GPG, Java, Spring, Gradle, G1 GC, JVM, Docker, Bash, Unix, DL4J, DL, RNN, CNN, NN, ML, Neo4J, DAO, Bolt, HTTP, REST, SSO, JWK, JWT, JSON, JavaScript, React, CSS, HTML, AEI
Fusek Michal, Ing., Ph.D. (DMAT FEEC BUT), člen
Martínek Tomáš, doc. Ing., Ph.D. (DCSY FIT BUT), člen
Matoušek Petr, doc. Ing., Ph.D., M.A. (DIFS FIT BUT), člen
Rogalewicz Adam, doc. Mgr., Ph.D. (DITS FIT BUT), člen
@bachelorsthesis{FITBT22278, author = "Tom\'{a}\v{s} Dole\v{z}al", type = "Bachelor's thesis", title = "Aplikace s rozpozn\'{a}v\'{a}n\'{i}m j\'{i}dla", school = "Brno University of Technology, Faculty of Information Technology", year = 2019, location = "Brno, CZ", language = "czech", url = "https://www.fit.vut.cz/study/thesis/22278/" }