An improved algorithm for sesame seedling and weed detection based on YOLOV7
by Gan Yu; Huimin Sun; Zhuolei Xiao; Chaoyue Dai
International Journal of Wireless and Mobile Computing (IJWMC), Vol. 26, No. 3, 2024

Abstract: Sesame is a widely cultivated oilseed and edible crop with significant economic and nutritional value. Weed infestation is a primary factor that hinders sesame growth, and effective weed control is crucial for optimal sesame yield. Accurate identification of sesame seedlings and weeds can guide intelligent devices to improve weed control. However, low recognition accuracy and high miss-detection rates persist as major issues. To address these challenges, we propose the YOLOV7-G algorithm, which builds upon the YOLOV7 baseline network. Our approach integrates the SimAM attention mechanism in the feature extraction structure to focus the model on the morphological features of weeds and incorporates the C3 module into the backbone network to increase the perceptual field range. We also employ the SPPFCSPC module in the Neck part to replace convolutional kernels of different sizes with stacked 5×5 convolutional kernels to reduce computational effort while maintaining the original perceptual field. Finally, we utilise the Focal-SIoU loss function to improve the regression accuracy of the prediction frame. Experimental results demonstrate that the YOLOV7-G algorithm out-performs YOLOV7, achieving a 17% increase in average precision a 5.1% increase in mAP@0.5.

Online publication date: Fri, 05-Apr-2024

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 Wireless and Mobile Computing (IJWMC):
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