Irfan Fajar Muttaqin, Riza Arifudin

Fruit Freshness Detection Using Android-Based Transfer Learning MobileNetV2

Introduction

Fruit freshness detection using android-based transfer learning mobilenetv2. Android app for fruit freshness detection using MobileNetV2 transfer learning. Achieves 99.62% accuracy for apples, bananas, oranges, aiding consumer choice.

Abstract

Abstract. Fruit is an important part of the source of food nutrition in humans. Fruit freshness is one of the most important factors in selecting fruit that is suitable for consumption. Fruit freshness is also an important factor in determining the price of fruit in the market. So it is very necessary to detect fruit freshness which can be done by machine. Take apples, bananas, and oranges as samples. The machine learning algorithm used in this study uses MobileNetV2 with transfer learning techniques. MobileNetV2 introduces many new ideas aimed at reducing the number of parameters to make it more efficient to run on mobile devices and achieve high classification accuracy. Transfer learning is used so that data does not need training from the start, so it only takes several networks from MobileNetV2 that have previously been trained and then retrained with a different purpose to improve accuracy results. Then the models that have been created are inserted into the application using Android Studio. Software testing is done through black box testing. Purpose: The purpose of this research is to design a machine-learning model to detect fruit freshness and then apply it to application Android smartphones. Methods/Study design/approach: The algorithm used in this study uses MobileNetV2 with transfer learning techniques. Models that have been created are inserted into the application using Android Studio. Result/Findings: The training results using MobileNetV2 transfer learning obtained an accuracy of 99.62% and the loss results obtained were 0.34%. The results of the application after testing using the black box testing method required improvements to the application and the machine learning model so that it can run optimally. Novelty/Originality/Value: Machine learning models that have been created using transfer learning MobileNetV2 are applied to Android applications so that they can be used by the public.

Review

This paper presents an Android-based system for detecting fruit freshness, specifically for apples, bananas, and oranges, utilizing the MobileNetV2 architecture with transfer learning. The authors aim to leverage machine learning to address the important factors of fruit suitability for consumption and market pricing. A key strength reported is the impressive training accuracy of 99.62% and a low loss of 0.34%, suggesting strong classification capabilities. The initiative to deploy this model into an Android application is commendable, demonstrating a practical approach to make such technology accessible to the public, aligning with the goal of creating a user-friendly tool. Despite the promising accuracy, the abstract raises several questions regarding the methodology and evaluation that limit a thorough assessment. Details about the dataset—its size, diversity, acquisition method, and how "freshness" was objectively defined and labeled for training—are critically absent. The description of using "several networks from MobileNetV2" for transfer learning is vague and lacks specificity regarding the fine-tuning process. Furthermore, while a high training accuracy is reported, there is no mention of independent test set performance, cross-validation, or a comparison to baseline models, which are crucial for validating the generalizability of the model. The statement that black box testing "required improvements to the application and the machine learning model so that it can run optimally" significantly undermines the reported high accuracy, implying the system is not yet robust for real-world deployment. In conclusion, the work presents a potentially valuable application of machine learning for fruit freshness detection, with the ambitious goal of mobile deployment. However, to fully evaluate its merits, the manuscript would need substantial expansion on its methodological details, including comprehensive descriptions of the dataset, explicit definitions of "freshness," and a more rigorous evaluation of the model's performance on unseen data. Addressing the inconsistencies between the reported high accuracy and the necessity for further improvements, along with clarifying the novelty beyond standard transfer learning application, would significantly strengthen the paper and its contribution to the field.

Full Text

Comments

You need to be logged in to post a comment.