Indexed by:
Abstract:
Rapid development of telecommunication technology in China has led to a prosperous market of smart phones, as well as an increase number of used phones. Nevertheless, there are key factors affecting the used phone recycling, one of which is the phone color. To realize an accurate automatic color recognition of used phones to enhance the recycling process, a high-dimensional spatial color conversion deep convolutional neural network (HSCCNet) is proposed in this paper. First, we established a common dataset for the field of used electronic devices. Second, the phone color is converted to the high-dimensional space of hue, saturation and value (HSV), which generates richer expressions of color features and improves the model sensitivity. Finally, a deep convolutional structure for HSV features is designed, where color feature conversion are implemented, resulting in enhanced color feature expressions. Promising results are obtained through the comparison between the proposed HSCCNet and the state-of-the-art models. © 2022
Keyword:
Reprint Author's Address:
Email:
Source :
Resources, Conservation and Recycling
ISSN: 0921-3449
Year: 2022
Volume: 187
1 3 . 2
JCR@2022
1 3 . 2 0 0
JCR@2022
ESI Discipline: ENVIRONMENT/ECOLOGY;
ESI HC Threshold:47
JCR Journal Grade:1
CAS Journal Grade:1
Cited Count:
SCOPUS Cited Count: 5
ESI Highly Cited Papers on the List: 0 Unfold All
WanFang Cited Count:
Chinese Cited Count:
30 Days PV: 1
Affiliated Colleges: