Convolutional neural network based detection and judgement of environmental obstacle in vehicle operation. Issue 2 (10th April 2019)

• Record Type: Journal Article
• Authors: Qi, Guanqiu; Wang, Huan; Haner, Matthew; Weng, Chenjie; Chen, Sixin; Zhu, Zhiqin
• Is Part Of: CAAI transactions on intelligence technology. Volume 4:Issue 2(2019)
• Page Start: 80
• Page End: 91
• Publication Date: 2019-04-10
• Subjects: convolution -- learning (artificial intelligence) -- neural nets -- object recognition; calculation efficiency -- traditional deep learning solutions -- unit work -- central processing unit -- deep convolutional neural network -- target recognition solution -- morphology -- maximum difference method -- interest extraction method -- extensive processing power -- high recognition accuracy -- deep‐learning based recognition techniques -- resource intensive -- vehicle automation -- real‐time obstacle recognition -- vehicle operation -- environmental obstacle -- judgement -- detection; C5260B Computer vision and image processing techniques -- C5290 Neural computing techniques -- C6170K Knowledge engineering techniques; Artificial intelligence -- Periodicals; Computer science -- Periodicals; Artificial intelligence; Computer science; Electronic journals; Periodicals; 006.305
• DOI: 10.1049/trit.2018.1045
• Languages: English
• ISSNs: 2468-6557

Abstract

Abstract : Precise real‐time obstacle recognition is both vital to vehicle automation and extremely resource intensive. Current deep‐learning based recognition techniques generally reach high recognition accuracy, but require extensive processing power. This study proposes a region of interest extraction method based on the maximum difference method and morphology, and a target recognition solution created with a deep convolutional neural network. In the proposed solution, the central processing unit and graphics processing unit work collaboratively. Compared with traditional deep learning solutions, the proposed solution decreases the complexity of algorithm, and improves both calculation efficiency and recognition accuracy. Overall it achieves a good balance between accuracy and computation.