Hand-drawn electronic component recognition using deep learning algorithm
by Haiyan Wang; Tianhong Pan; Mian Khuram Ahsan
International Journal of Computer Applications in Technology (IJCAT), Vol. 62, No. 1, 2020

Abstract: Hand-drawn circuit recognition plays an increasingly important role in circuit design work and electrical knowledge teaching. Hand-drawn electronic component recognition is an indispensable part of hand-drawn circuit recognition. Accurate electronic component recognition ensures accurate circuit recognition. In this paper, a hand-drawn electronic component recognition method using a Convolutional Neural Network (CNN) and a softmax classifier is proposed. The CNN composed of a convolutional layer, an activation layer and an average-pooling layer is designed to extract features of a hand-drawn electronic component image. The kernel function for the CNN is obtained by a sparse auto-encoder method. A softmax classifier is trained for classification based on the features extracted by the CNN. The recognition method can identify rotating electronic components because of the added rotated image and achieve 95% recognition accuracy.

Online publication date: Mon, 02-Dec-2019

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computer Applications in Technology (IJCAT):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com