Development of Tomato Leaf Disease Detection using Single Shot Detector (SSD) Mobilenet V2

  • Stanley Glenn E. Brucal School of Engineering, Asia Pacific College, Philippines
  • Luigi Carlo M. De Jesus School of Engineering, Asia Pacific College, Philippines
  • Jex O. De Los Santos School of Engineering, Asia Pacific College, Philippines
  • Mariel Joy V. Mendoza School of Engineering, Asia Pacific College, Philippines
  • Khyrstelle E. Harion School of Engineering, Asia Pacific College, Philippines
  • Guiliane Altaire S. Reyes School of Engineering, Asia Pacific College, Philippines
  • Dominador S. Nevalasca School of Engineering, Asia Pacific College, Philippines
  • Jv Kay C. Reyes School of Engineering, Asia Pacific College, Philippines

Abstract

Purpose – To create a software prototype for the tomato leaf disease detection model to identify the tomato leaf condition and detect and identify the disease present in it.

Methodology – Using the TensorFlow 2 Object Detection API, the object detection model used is the Single Shot Detector (SSD) MobileNetV2 Object Detection model. The feature extractor used is the pre-trained TF2 MobileNetV2 model with the ImageNet dataset providing trained weights that allows feature extraction. Combining the pre-trained TF2 MobileNetV2 and Convolutional Neural Network (CNN) for SSD, the result object localization and image classification with SSD, and feature extractor pre-trained model.

Result – When training the model, at the 1300th step out of 6000 steps, the learning rate spiked from 0 to 0.7999. It then stabilized from 0.7999 and gradually decreased to 0.7796. After training, the total loss of the model is 46.95% for evaluation and 45.32% for training results. The average recall of the model is 94.51%, and the map of the model is 92.45%.

Conclusion –  Using TF2 MobileNetV2 as the pre-trained feature extractor and, adding it with SSD layers was developed to detect the tomato leaf condition. The model can recognize a healthy tomato leaf, a tomato leaf with bacterial spot, and a tomato leaf with yellow leaf curl virus correctly 94.51% of the time, with a mean average precision of the model at 92.45%.

Recommendation – The prototype can be improved by increasing the number of tomato leaf image datasets and expanding the types of tomato leaf diseases. A mobile application for real-time plant detection can improve its user-friendliness and usability.

Practical Implications – The prototype can help farmers and consumers identify tomato leaf diseases, enabling them to introduce early cures, mitigation, or prevention of tomato plant diseases. Hence, help increase in production and harvest of healthy tomato fruits.

Author Biographies

Stanley Glenn E. Brucal, School of Engineering, Asia Pacific College, Philippines

Engr. Stanley Glenn E. Brucal is a BS in Electronics Engineering graduate from Adamson University (2003) and a Master of Engineering major in Electronics and Communications Engineering from De La Salle University (2006). He is a Professional Electronics Engineer (PECE), an ASEAN Engineer (ASEAN Eng.), an ASEAN Chartered Professional Engineer (ACPE), and currently the Registrar of Asia Pacific College.

Luigi Carlo M. De Jesus, School of Engineering, Asia Pacific College, Philippines

Engr. Luigi Carlo M. De Jesus is a Master of Engineering Major in Computer Engineering graduate of Asia Pacific College. He is a licensed Electronics Engineer, Electronics Technician (2nd Placer), and the current Engineering and Science Laboratory Office (ESLO) Head of Asia Pacific College - School of Engineering.

Jex O. De Los Santos, School of Engineering, Asia Pacific College, Philippines

Jex O. De Los Santos is an SM Scholar student at Asia Pacific College pursuing a Bachelor of Science in Computer Engineering. He oversees the development and quality assurance of the model, which assisted the researcher in determining whether the project's functionalities and objectives were satisfied.

Mariel Joy V. Mendoza, School of Engineering, Asia Pacific College, Philippines
Mariel Joy V. Mendoza is an AFP-OLC Scholar student at Asia Pacific College pursuing a Bachelor of Science in Computer Engineering. She is in charge of overseeing the complete team's operations and deliverables, as well as contributing to the project's in-depth research, testing, and evaluation.
Khyrstelle E. Harion, School of Engineering, Asia Pacific College, Philippines

Khrystelle E. Harion is a student at Asia Pacific College pursuing a Bachelor of Science in Computer Engineering. She oversees the flow of the study's functions and means.

Guiliane Altaire S. Reyes, School of Engineering, Asia Pacific College, Philippines

Guilian Altaire S. Reyes is a student at Asia Pacific College pursuing a Bachelor of Science in Computer Engineering. She assisted with the training of the model and the documentation of the project.

Dominador S. Nevalasca, School of Engineering, Asia Pacific College, Philippines

Dominador S. Nevalasca is a student at Asia Pacific College pursuing a Bachelor of Science in Computer Engineering. He oversees gathering the datasets and he also assisted with the documentation of the project.

Published
2023-04-13
How to Cite
BRUCAL, Stanley Glenn E. et al. Development of Tomato Leaf Disease Detection using Single Shot Detector (SSD) Mobilenet V2. International Journal of Computing Sciences Research, [S.l.], v. 7, p. 1857-1869, apr. 2023. ISSN 2546-115X. Available at: <//stepacademic.net/ijcsr/article/view/405>. Date accessed: 24 july 2024.
Section
Special Issue: IRCCETE 2023

Most read articles by the same author(s)