|
|
Impact of color models on performance of convolutional neural networks
Šimunský, Martin ; Doležel, Petr (referee) ; Škrabánek, Pavel (advisor)
Current knowledge about impact of colour models on performance of convolutional neural network is investigated in the first part of this thesis. The experiment based on obtained knowledge is conducted in the second part. Six colour models HSV, CIE 1931 XYZ, CIE 1976 L*a*b*, YIQ a YCbCr and deep convolutional neural network ResNet-101 are used. RGB colour model achieved the highest classification accuracy, whereas HSV color model has the lowest accuracy in this experiment.
|
|
|
Static image compression techniques
Jirounek, Matěj ; Šmirg, Ondřej (referee) ; Krajsa, Ondřej (advisor)
Bachelor’s thesis dissert on nowadays used compression technologies static image, about their keystones, advantages and disadvantages in field of application and their reciprocal comparison. Work is divided to the seven capitols, second of which handles indicative allocation data compression, third about lossless compression and their main representative and fourth chapter is about less compression picture with her advantages and disadvantages as well as about colour models. In fifth chapter there are summarized used criteria for classification picture and in sixth chapter is shownd implementation programme in environment MATLAB.
|
|
|
Fire Detection in Images and Video
Poledník, Tomáš ; Juránková, Markéta (referee) ; Přibyl, Bronislav (advisor)
This bachelor's thesis deals with fire detection in image and video. Attention is paid to the existing methods and known approaches to fire detection. Based on the analysis of the described methods, a fire detector, monitoring the attributes of its colour and motion, is designed, implemented and tested. The implemented detector achieves true positive detection rate of up to 90%.
|
|
|
Impact of color models on performance of convolutional neural networks
Šimunský, Martin ; Doležel, Petr (referee) ; Škrabánek, Pavel (advisor)
Current knowledge about impact of colour models on performance of convolutional neural network is investigated in the first part of this thesis. The experiment based on obtained knowledge is conducted in the second part. Six colour models HSV, CIE 1931 XYZ, CIE 1976 L*a*b*, YIQ a YCbCr and deep convolutional neural network ResNet-101 are used. RGB colour model achieved the highest classification accuracy, whereas HSV color model has the lowest accuracy in this experiment.
|
|
|
Fire Detection in Images and Video
Poledník, Tomáš ; Juránková, Markéta (referee) ; Přibyl, Bronislav (advisor)
This bachelor's thesis deals with fire detection in image and video. Attention is paid to the existing methods and known approaches to fire detection. Based on the analysis of the described methods, a fire detector, monitoring the attributes of its colour and motion, is designed, implemented and tested. The implemented detector achieves true positive detection rate of up to 90%.
|
|
|
Static image compression techniques
Jirounek, Matěj ; Šmirg, Ondřej (referee) ; Krajsa, Ondřej (advisor)
Bachelor’s thesis dissert on nowadays used compression technologies static image, about their keystones, advantages and disadvantages in field of application and their reciprocal comparison. Work is divided to the seven capitols, second of which handles indicative allocation data compression, third about lossless compression and their main representative and fourth chapter is about less compression picture with her advantages and disadvantages as well as about colour models. In fifth chapter there are summarized used criteria for classification picture and in sixth chapter is shownd implementation programme in environment MATLAB.
|
guest ::
