Vehicle type classification based on convolutional neural network
| Autor: | Li-Dong Zhang, Wen-Xing Zhu, Yanjun Chen, Donghui Yao |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Normalization (statistics)
Pixel Artificial neural network Computer science business.industry Feature extraction ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Pattern recognition 02 engineering and technology 010501 environmental sciences 01 natural sciences Convolutional neural network Statistical classification Kernel (image processing) Softmax function 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence business Classifier (UML) 0105 earth and related environmental sciences |
| Zdroj: | 2017 Chinese Automation Congress (CAC). |
| DOI: | 10.1109/cac.2017.8243078 |
| Popis: | This paper aims to develop a framework for vehicle type classification using convolutional neural network based on vehicle rear view images. Compared with the extraction of the appearance features from vehicle side view and frontal view images, there has been relatively little research on vehicle type classification by using vehicle rear view images' information. The vehicle rear view images are detected from the traffic video frames. The image is normalized to the fixed size of 32∗32 pixels, and then through the removing-mean operation, the normalized image is sent into the trained network. As the last layer of the neural network, the softmax classifier can output the probability that the input image belongs to a certain type of vehicle. The experimental result shows that compared with traditional pattern recognition method, our algorithm performs better and eventually achieves the accuracy of 97.88% in vehicle type classification. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|