Jiawei Lian, Junhong He, Yun Niu and Tianze Wang
The current popular image processing technologies based on convolutional neural network have the characteristics of large computation, high storage cost and low accuracy for tiny…
Abstract
Purpose
The current popular image processing technologies based on convolutional neural network have the characteristics of large computation, high storage cost and low accuracy for tiny defect detection, which is contrary to the high real-time and accuracy, limited computing resources and storage required by industrial applications. Therefore, an improved YOLOv4 named as YOLOv4-Defect is proposed aim to solve the above problems.
Design/methodology/approach
On the one hand, this study performs multi-dimensional compression processing on the feature extraction network of YOLOv4 to simplify the model and improve the feature extraction ability of the model through knowledge distillation. On the other hand, a prediction scale with more detailed receptive field is added to optimize the model structure, which can improve the detection performance for tiny defects.
Findings
The effectiveness of the method is verified by public data sets NEU-CLS and DAGM 2007, and the steel ingot data set collected in the actual industrial field. The experimental results demonstrated that the proposed YOLOv4-Defect method can greatly improve the recognition efficiency and accuracy and reduce the size and computation consumption of the model.
Originality/value
This paper proposed an improved YOLOv4 named as YOLOv4-Defect for the detection of surface defect, which is conducive to application in various industrial scenarios with limited storage and computing resources, and meets the requirements of high real-time and precision.
Details
Keywords
Durairaj Maheswaran, Cathy Yi Chen and Junhong He
Purpose – Extensive research in the area of consumer behavior has documented the “Country of Origin Effect,” which identifies country of origin as an important decision variable…
Abstract
Purpose – Extensive research in the area of consumer behavior has documented the “Country of Origin Effect,” which identifies country of origin as an important decision variable in evaluating products and services. Past research has mostly assumed that country of origin effect is driven by the performance of the products originating in that country. However, consumers can also form opinions about countries based on exposure to information that is unrelated to the product and may have roots in macro factors such as history, culture, and politics. These emotions, while extraneous to the product, can also influence product evaluations along with performance-related country information.Design/methodology/approach – This review examines research addressing both performance and emotional perceptions related to country of origin.Findings – This review presents an integrating framework termed “Nation Equity” to systematically understand and examine the influence of various dimensions of country of origin on consumer decision making.
Details
Keywords
Niu Jian, Xiao Junhong, Wang Zhongfeng and He Lanxiang
Web-based course assessment is a new thing at China's Open University – China Radio and Television Universities (China RTVUs). This article reports an innovative pilot study in…
Abstract
Web-based course assessment is a new thing at China's Open University – China Radio and Television Universities (China RTVUs). This article reports an innovative pilot study in this research area. The experimental course for integrated web-based assessment in this study is Advanced English Writing, which is a compulsory course in the B.A. English programme at China Central Radio and Television University (CCRTVU). The study started in March 2005 and it is still in progress at the moment. This article first describes the webbased assessment design of the course and the implementation procedures. Then it moves on to report some initial feedback from the student participants on the pilot study. The article ends with a tentative plan for further actions based on the current study.