An intelligent factory prediction model based on big data related analysis and machine learning

Zhang, Yuelong; Diao, Junjie; Xue, Hui; Hong, Wanchen; Yin, Xiangquan

doi:10.1117/12.2635377

Cited by 1 publication

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Following automated and digital factories, smart factories featuring the Internet of Things, safety monitoring systems, storage equipment management systems and storage operation information systems have become a development goal and direction for modern manufacturing enterprises [1][2][3][4][5][6][7]. In the actual operation of smart factories, warehousing and distribution have become the main factors restricting their development.…”

Section: Introductionmentioning

confidence: 99%

Image Information Recognition and Classification of Warehoused Goods in Intelligent Logistics Based on Machine Vision Technology

He¹,

Wang²,

Liu³

2022

View full text Add to dashboard Cite

To sort out warehouse management problems in smart factories, smart warehousing and in-plant smart distribution systems are needed to achieve the goal of lean logistics and distribution in smart factories. There are still some pressing problems in the research on images of warehoused goods in intelligent logistics. For example, a solution hasn’t been found yet to recognise multiple types of warehoused goods in different shapes and colours; static vision image processing solutions have a poor performance in optimising recognition speed and classification accuracy. In response, this paper unveils a study on the image information recognition and classification of warehoused goods in intelligent logistics based on machine vision technology. It presents a process related to warehouse management in intelligent logistics and a corresponding system architecture. It also constructs a YOLOv3 model for the image information recognition and classification of warehoused goods in intelligent logistics. The paper elaborates on the prior box settings and loss function correction methods, and finishes optimising the YOLOv3 model. Experimental results verified the effectiveness of the constructed model.

show abstract