|
|
Holistic License Plate Recognition Based on Convolution Neural Networks
Morbitzer, Dušan ; Juránek, Roman (referee) ; Špaňhel, Jakub (advisor)
The goal of this work is to create a model of neural network for holistic recognition of license plates, focused on accuracy and shortening of the learning process. The model was implemented as a union of convolutional neural network for extraction of deep features of a plate and Bidirectional LSTM with CTC. The trained model was compared to another implementation using a holistic approach, that was trained on the same dataset. My design of the network achieved better results in recognition on a dataset, which is different from the training one, with an error rate of 8.3 %.
|
|
|
Re-Identification of Vehicles by License Plate Recognition
Špaňhel, Jakub ; Juránková, Markéta (referee) ; Herout, Adam (advisor)
This thesis aims at proposing vehicle license plate detection and recognition algorithms, suitable for vehicle re-identification. Simple urban traffic analysis system is also proposed. Multiple stages of this system was developed and tested. Specifically - vehicle detection, license plate detection and recognition. Vehicle detection is based on background substraction method, which results in an average hit rate of ~92%. License plate detection is done by cascade classifiers and achieves an average hit rate of 81.92% and precision rate of 94.42%. License plate recognition based on Template matching results in an average precission rate of 60.55%. Therefore the new license plate recognition method based on license plate scanning using the sliding window principle and neural network recognition was introduced. Neural network achieves a precision rate of 64.47% for five input features. Low precision rate of neural network is caused by small amount of training sample for some specific license plate characters.
|
|
|
System for Automatic Parking Access Based on License Plate Recognition
Václavek, Patrik ; Sochor, Jakub (referee) ; Špaňhel, Jakub (advisor)
Goal of this thesis was to design and implement system operating in real time, which manages to detect incoming vehicle to the car park terminal, recognize its licence plate and automatically decide on its admission. System uses the Gaussian Mixture Model algorithm for detection of incoming vehicle. For reliable localization of licence plate are used two methods, the first one uses of extraction of Maximally Stable Extremal Regions (MSERs), the second one uses of Top-Hat transformation. Support Vector Machine (SVM) algorithm is used to decide, whether is the found area a licence plate. Character classification is performed using artificial neural network. For implementation was used library OpenCV. Thanks to optimalization is the extraction of MSERs accelerated up to seven times. The accomplished success rate in case of licence plate localization is 92,47% and in case of classification of characters is 90,03%.
|
|
|
Holistic License Plate Recognition Based on Convolution Neural Networks
Morbitzer, Dušan ; Juránek, Roman (referee) ; Špaňhel, Jakub (advisor)
The goal of this work is to create a model of neural network for holistic recognition of license plates, focused on accuracy and shortening of the learning process. The model was implemented as a union of convolutional neural network for extraction of deep features of a plate and Bidirectional LSTM with CTC. The trained model was compared to another implementation using a holistic approach, that was trained on the same dataset. My design of the network achieved better results in recognition on a dataset, which is different from the training one, with an error rate of 8.3 %.
|
|
|
Re-Identification of Vehicles by License Plate Recognition
Špaňhel, Jakub ; Juránková, Markéta (referee) ; Herout, Adam (advisor)
This thesis aims at proposing vehicle license plate detection and recognition algorithms, suitable for vehicle re-identification. Simple urban traffic analysis system is also proposed. Multiple stages of this system was developed and tested. Specifically - vehicle detection, license plate detection and recognition. Vehicle detection is based on background substraction method, which results in an average hit rate of ~92%. License plate detection is done by cascade classifiers and achieves an average hit rate of 81.92% and precision rate of 94.42%. License plate recognition based on Template matching results in an average precission rate of 60.55%. Therefore the new license plate recognition method based on license plate scanning using the sliding window principle and neural network recognition was introduced. Neural network achieves a precision rate of 64.47% for five input features. Low precision rate of neural network is caused by small amount of training sample for some specific license plate characters.
|
|
|
System for Automatic Parking Access Based on License Plate Recognition
Václavek, Patrik ; Sochor, Jakub (referee) ; Špaňhel, Jakub (advisor)
Goal of this thesis was to design and implement system operating in real time, which manages to detect incoming vehicle to the car park terminal, recognize its licence plate and automatically decide on its admission. System uses the Gaussian Mixture Model algorithm for detection of incoming vehicle. For reliable localization of licence plate are used two methods, the first one uses of extraction of Maximally Stable Extremal Regions (MSERs), the second one uses of Top-Hat transformation. Support Vector Machine (SVM) algorithm is used to decide, whether is the found area a licence plate. Character classification is performed using artificial neural network. For implementation was used library OpenCV. Thanks to optimalization is the extraction of MSERs accelerated up to seven times. The accomplished success rate in case of licence plate localization is 92,47% and in case of classification of characters is 90,03%.
|
guest ::
