Abstract
To encourage people to save energy, subcompact cars have several benefits of discount on parking or toll road charge. However, manual classification of the subcompact car is highly labor intensive. To solve this problem, automatic vehicle classification systems are good candidates. Since a general pattern-based classification technique can not successfully recognize the ambiguous features of a vehicle, we present a new multi-resolution convolutional neural network (CNN) and stochastic orthogonal learning method to train the network. We first extract the region of a bonnet in the vehicle image. Next, both extracted and input image are engaged to low and high resolution layers in the CNN model. The proposed network is then optimized based on stochastic orthogonality. We also built a novel subcompact vehicle dataset that will be open for a public use. Experimental results show that the proposed model outperforms state-of-the-art approaches in term of accuracy, which means that the proposed method can efficiently classify the ambiguous features between subcompact and non-subcompact vehicles.
Highlights
Subcompact cars are defined by the engine displacement, width, and height under1000 cc, 1.6 m, and 2.0 m, respectively
To classify the visual feature in images, Dalal et al extracted the histogram of oriented gradients (HOG) and classify the HOG using support vector machine (SVM) [3]
Proposed by Dalal et al [3], multi-channel correlation filter (MCCF) combined HOG recognition (MCCF + HOG + SV M) [15], multi-resolution image based HOG recognition (Retinex + MCCF + HOG + SV M), deep neural network based Huttunen’s method (DNN) [8], convolutional neural network (CNN) with 16 layers proposed by Simonyan et al [9], retinex CNN Retinex + CNN, based on proposed pre-convolution layer (Retinex + MCCF + CNN), and the proposed multi-resolution network without orthogonal learning (Retinex + MCCF + MRN)
Summary
Subcompact cars are defined by the engine displacement, width, and height under. 1000 cc, 1.6 m, and 2.0 m, respectively To satisfy these specifications, the subcompact car has a unique shape such as shorter-bonnet and hatchback. There are various environmental benefits because the subcompact cars have a small displacement engine and a light weight. Vehicle classification methods can be classified into two approaches: one uses infrared sensors to measure physical dimensions of a vehicle such as length, height, and width. The other uses a single camera and image processing algorithms to recognize the visual characteristics of vehicles [1,2]. Despite of accuracy and robustness, the infrared sensor-based system is too expensive to be installed in many places.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
