|
|
Demonstration tasks of computer vision on a CNC milling machine
Duda, Pavel ; Honec, Peter (referee) ; Janáková, Ilona (advisor)
This bachelor thesis deals with the integration of computer vision into CNC machines in order to demonstrate the practical application of this technology. The task is to design several demonstration tasks to test and implement the various computer vision algorithms and subsequent data processing to control a CNC milling machine. In conclusion, we find an evaluation of the results and possible alternative steps to fulfill the problem.
|
|
|
Automation of video extensometers using artificial intelligence
Leinweber, Vít ; Adamec, Tomáš (referee) ; Ščerba, Bořek (advisor)
This thesis deals with the automation of the use of video extensometers using artificial intelligence methods, specifically the implementation of a suitable algorithm for the recognition of the type of sample to be measured and the placement of a suitable software tool in the correct position on the sample in real-time. The theoretical part of the thesis describes digital image correlation, computer vision with a focus on object recognition, and machine learning with a focus on deep convolutional neural networks and their architectures for object detection. Based on the findings from the theoretical part of the work, the YOLOv8 algorithm is chosen as the most suitable algorithm for classifying the type of sample and determining its location in real-time. In the practical part of the thesis, a dataset containing three types of samples is created using the database of video extensometer manufacturer. The dataset is extended with images containing samples to be detected and augmented. The selected algorithm is trained, optimized, and tested using the created dataset. A genetic algorithm and a random search of the hyperparameter space are used in the optimization process. The best trained models of the YOLOv8 algorithm are compared with each other on two test sets and the best one is selected. Furthermore, the work with this algorithm as a detector integrated into the corresponding software for working with video extensometers is described. Finally, a method for correcting the effect of potential misalignment of samples in the images entering the detector is proposed.
|
|
|
Adversarial Attacks on AI Algorithms and Their Prevention
Gregorová, Jana ; Vaško, Marek (referee) ; Herout, Adam (advisor)
Bezpečnost AI a útoky na umělou inteligenci představují komplexní a dosud nedostatečně prozkoumanou problematiku. Cílem této práce je nabídnout ucelený přehled klíčových metod a možných vysvětlení útoků na AI a obran proti nim, aby se toto téma stalo přístupnějším a srozumitelnějším pro širší publikum, a tím usnadnit hlubší zkoumání a porozumění těmto útokům. Tato práce zahrnuje výběr metod pro vysvětlení jednotlivých klasifikačních rozhodnutí klasifikátorů hlubokého učení (Explainable AI: XAI) a jejich aplikaci při analýze rozhodovacího procesu klasifikátorů během útoků. K usnadnění vytváření dalších experimentů, monitorování útoků na AI a hledání možných vysvětlení byl navíc vyvinut skript, který tento proces zjednodušuje. Tento skript je součástí této práce a je poskytnut na přiloženém médiu.
|
|
|
Identifikace člověka podle fotografie dlaně / hřbetu ruky
Štanga, Miroslav ; Vaško, Marek (referee) ; Herout, Adam (advisor)
This work focuses on using contrastive self-supervised learning method for creating model of deep learning intended for person recognition based on hand photographs. The paper outlines fundamentals of machine learning, utilized tools and dataset. The method was developed using PyTorch library. The proposed model draws inspiration from the SimCLR architecture and its use of contrastive representation learning. The proposed approach utilizes the triplet loss function for optimization. Then the optimization process is described and impact of individual hyperparameters on the model´s accuracy is compared. The resulting model was trained on 1696 hand photos and achieves 98% accuracy on validation set. The accuracy achieved using self-supervised methods is higher than the accuracy achieved using supervised methods.
|
|
|
Evaluation of Image Quality and Camera Setup
Ondris, Ladislav ; Fučík, Otto (referee) ; Zemčík, Pavel (advisor)
Tato práce si klade za cíl vyvinout základ pro univerzální fotoaparát schopný aktualizovat své parametry na základě pozorované scény. Tento přístup kombinuje metriky hodnocení kvality obrazu s rozpoznáváním scény. Byla shromážděna sada metrik, jako jsou ty používané k vyhodnocení kontrastu a ostrosti. Kromě toho byl vyvinut model strojového učení pro rozpoznávání scény, který má sloužit jako základ pro výběr vhodných parametrů fotoaparátu přizpůsobených konkrétní scéně. Práce demonstruje praktické využití metrik k optimalizaci vybraných parametrů obrazového procesoru a k detekci optických aberací.
|
|
|
Comparison of Methods for Image Inpainting based on Deep Learning
Rajsigl, Tomáš ; Herout, Adam (referee) ; Španěl, Michal (advisor)
This bachelor thesis aims to compare deep learning methods and approaches for image inpainting using quantitative metrics like PSNR, SSIM, and LPIPS. Moreover, a user study has also been carried out for further subjective assessment. For the purposes of this comparison, four GAN-based neural networks were used. The first network, AOT-GAN, represents a benchmark against which the proposed architecture and its modifications were compared. In the experiments, a variant of the proposed method achieved a 29% improvement against AOT-GAN in images with small missing regions. This claim is also supported by the results of the user study where this method was ranked as the best. As a result of this thesis, a small dataset specifically for the evaluation of image inpainting in the context of object removal was created. Real-world applications of these methods are demonstrated through a web application.
|
|
|
Surface defect detection of metal parts based on neural networks
Hadwiger, Tomáš ; Jonák, Martin (referee) ; Ježek, Štěpán (advisor)
The goal of this thesis is focused on surface anomaly detection on metal parts. The goal was to implement different neural network architectures using the method CutPaste and compare them on three different datasets: MVTec AD, MPDD, MPDD2. For the object classes of the dataset MVTec AD the most accurate architecture turned out to be ResNet-18 with average precision of 84,45 AUROC, for the materials it was the EfficientNet architecture with average precision of 87,22 AUROC. For the MPDD and MPDD2 datasets, the most accurate architecture was ResNet50 with average precision of 88,64 AUROC and 61,10 AUROC respectively. Based on the measure values, the most difficult dataset for anomaly detection turned out to be MPDD2.
|
|
|
Design and implementation of an obstacle avoidance method in an outdoor environment for a mobile robot
Fargač, Tomáš ; Králík, Jan (referee) ; Věchet, Stanislav (advisor)
This thesis focuses on investigating the usability of the optical flow method in image processing. Firstly, this method is introduced theoretically, followed by its mathematical derivation. Subsequently, the idea of implementing it into decision-making algorithms and potential areas of application is presented. The thesis also elaborates on suitable environments for such applications in both virtual and real worlds. The practical part demonstrates the step-by-step development process and the refinement of working with this method and its outputs. The work utilizes the Matlab programming environment and detailed work at the level of individual components in this programming language, enriched with auxiliary toolboxes, especially from the field of computer vision. The entire research is summarized clearly at the end, and all undertaken steps are depicted in a flowchart. Finally, all explored approaches with their strengths and weaknesses, identified throughout the process, are clearly presented.
|
|
|
Road and path segmentation in images for autonomous driving scenario
Janíček, Ondřej ; Cihlář, Miloš (referee) ; Svědiroh, Stanislav (advisor)
This bachelor's thesis deals with the topic of segmentation of roads and paths for the purposes of autonomous driving. In the theoretical part, it deals with computer vision, simple segmentation methods, and practical solutions to the problem using convolutional neural networks and classical methods. In the practical part, the work deals with the collection of test data, the selection of a suitable programming language, and the selection of suitable libraries. Subsequently, the procedure for programming our own solution will be presented. Here it starts with pre-processing to convert the image into a grayscale image and filtering the noise, then finding the edges in the image using the Canny edge detector, followed by the definition of the region of interest, with the subsequent Hough transform to detect the straight lines in the image, and in the last stage, filtering the horizontal lines and averaging the remaining lines. At the end of the thesis, the results of the presented solution are compared with respect to robustness and computational complexity.
|
|
|
Detection of Material Surface Damage Based on a Photograph
Valko, Marek ; Malaník, Petr (referee) ; Dyk, Tomáš (advisor)
This thesis focuses on the detection of surface defects from photograph using computer vision techniques, convolutional neural networks and models for object detection such as Faster R-CNN and YOLO. Different methods of surface damage detection, image processing, neural networks and machine learning are described in detail. The thesis also compares the performance of these models in identification of surface defects on wood and steel using different augmentations of these datasets.
|
guest ::
