Stefano Polimena , Gianvito Pio , Maria Cefola , Michela Palumbo , Michelangelo Ceci , Giovanni Attolico
{"title":"A novel random forest-based approach for the non-destructive and explainable estimation of ammonia and chlorophyll in fresh-cut rocket leaves","authors":"Stefano Polimena , Gianvito Pio , Maria Cefola , Michela Palumbo , Michelangelo Ceci , Giovanni Attolico","doi":"10.1016/j.inpa.2024.09.002","DOIUrl":null,"url":null,"abstract":"<div><div>The perceived visual quality of fruits and vegetables plays a central role in the choices made by retail customers. Machine learning (ML) approaches based on image analysis have been recently proposed to overcome the poor efficiency and subjectivity of human visual evaluation as well as the expensiveness and destructiveness of physical and chemical methods that measure internal indicators. In this paper, we propose a ML method based on Random Forests for estimating the chlorophyll and ammonia contents (considered, in the literature, reliable indicators of product freshness) from images of fresh-cut rocket leaves. Our approach copes with specific issues raised by (i) the non-uniform distributions of ammonia and chlorophyll values and (ii) the need to provide insights into the features that produce a particular model outcome, aiming to enhance its trustworthiness. Our experiments, performed on real images of fresh-cut rocket leaves, proved that the proposed approach significantly outperforms 7 competitor methods, obtaining an improvement of the RSE results of 6.6% for the prediction of the ammonia and of 10.4% for the prediction of the chlorophyll over its best competitor. Moreover, a specific analysis of the explainability of the predictions showed that the learned models are based on reasonable features, empowering their acceptance in real-world applications.</div></div>","PeriodicalId":53443,"journal":{"name":"Information Processing in Agriculture","volume":"12 2","pages":"Pages 221-231"},"PeriodicalIF":7.7000,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Information Processing in Agriculture","FirstCategoryId":"1091","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2214317324000635","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AGRICULTURE, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
Abstract
The perceived visual quality of fruits and vegetables plays a central role in the choices made by retail customers. Machine learning (ML) approaches based on image analysis have been recently proposed to overcome the poor efficiency and subjectivity of human visual evaluation as well as the expensiveness and destructiveness of physical and chemical methods that measure internal indicators. In this paper, we propose a ML method based on Random Forests for estimating the chlorophyll and ammonia contents (considered, in the literature, reliable indicators of product freshness) from images of fresh-cut rocket leaves. Our approach copes with specific issues raised by (i) the non-uniform distributions of ammonia and chlorophyll values and (ii) the need to provide insights into the features that produce a particular model outcome, aiming to enhance its trustworthiness. Our experiments, performed on real images of fresh-cut rocket leaves, proved that the proposed approach significantly outperforms 7 competitor methods, obtaining an improvement of the RSE results of 6.6% for the prediction of the ammonia and of 10.4% for the prediction of the chlorophyll over its best competitor. Moreover, a specific analysis of the explainability of the predictions showed that the learned models are based on reasonable features, empowering their acceptance in real-world applications.
期刊介绍:
Information Processing in Agriculture (IPA) was established in 2013 and it encourages the development towards a science and technology of information processing in agriculture, through the following aims: • Promote the use of knowledge and methods from the information processing technologies in the agriculture; • Illustrate the experiences and publications of the institutes, universities and government, and also the profitable technologies on agriculture; • Provide opportunities and platform for exchanging knowledge, strategies and experiences among the researchers in information processing worldwide; • Promote and encourage interactions among agriculture Scientists, Meteorologists, Biologists (Pathologists/Entomologists) with IT Professionals and other stakeholders to develop and implement methods, techniques, tools, and issues related to information processing technology in agriculture; • Create and promote expert groups for development of agro-meteorological databases, crop and livestock modelling and applications for development of crop performance based decision support system. Topics of interest include, but are not limited to: • Smart Sensor and Wireless Sensor Network • Remote Sensing • Simulation, Optimization, Modeling and Automatic Control • Decision Support Systems, Intelligent Systems and Artificial Intelligence • Computer Vision and Image Processing • Inspection and Traceability for Food Quality • Precision Agriculture and Intelligent Instrument • The Internet of Things and Cloud Computing • Big Data and Data Mining