|
|
Recognition of Driving Lane Borders in Video from On-Board Camera
Fridrich, David ; Kohút, Jan (referee) ; Herout, Adam (advisor)
This paper talks about lane detection. Specifically custom generator of synthetic images, usage during training of neural networks, testing on convolutional neural network (CNN) UNet model and possibilities of extension of this model to SALMnet (Structure-Aware Lane Marking Detection Network) via addding SGCA module (semantic-guided channel attention) and PDC module (pyramid deformable convolution). Training results from synthetic datasets show very accurate results, reaching around 95\,\% in accuracy (even 99\,\% for easier images). Trainings with real datasets show lower accuracy, depending on the difficulty of the dataset itself. TuSimple has easier and clearer images and reaches about 62\,\%. CuLane is much more complex and results show accuracy around 37\,\%.
|
|
|
Support for Predictive Application Autoscaling on Kubernetes Platform
Fridrich, David ; Pavela, Jiří (referee) ; Rogalewicz, Adam (advisor)
The goal of this work is to create a new interface that will allow users to process collected metrics for scaling according to a formula (e.g. average value, mathematical equations, conditional statements) defined by a user. It also allows users to use an external interface for connecting KEDA to a component that defines its own scaling behavior, with which the user can achieve more complex solutions like automated predictive scaling of applications on Kubernetes platform. I solved the selected problems by modifying the KEDA core by implementing a new interface for scaling according to a custom formula with arithmetic and conditional expressions and the ability to connect a custom external remote method for calculating metrics using gRPC technology. The created solution provides a more flexible way to process metrics and also allows user to implement their own methods.
|
|
|
Recognition of Driving Lane Borders in Video from On-Board Camera
Fridrich, David ; Kohút, Jan (referee) ; Herout, Adam (advisor)
This paper talks about lane detection. Specifically custom generator of synthetic images, usage during training of neural networks, testing on convolutional neural network (CNN) UNet model and possibilities of extension of this model to SALMnet (Structure-Aware Lane Marking Detection Network) via addding SGCA module (semantic-guided channel attention) and PDC module (pyramid deformable convolution). Training results from synthetic datasets show very accurate results, reaching around 95\,\% in accuracy (even 99\,\% for easier images). Trainings with real datasets show lower accuracy, depending on the difficulty of the dataset itself. TuSimple has easier and clearer images and reaches about 62\,\%. CuLane is much more complex and results show accuracy around 37\,\%.
|
guest ::
RSS feed.