Facilitated machine learning for image-based fruit quality assessment

Knott, Manuel; Pérez-Cruz, Fernando; Defraeye, Thijs

doi:10.1016/j.jfoodeng.2022.111401

Cited by 21 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ML models with a smartphone camera can also be implemented as a low-cost technique to detect the quality and maturity of crops during storage and distribution (Cavallo et al ., 2019; Cho et al ., 2020; de Souza et al ., 2021; Garre et al ., 2020). ML is therefore seen as a viable option for the development and implementation of non-destructive evaluation as well as sorting and grading processes for varied cultivars (Dhakshina Kumar et al ., 2020; Knott et al ., 2023; Khaled et al ., 2022).…”

Section: Results and Analysismentioning

confidence: 99%

Implementation of information and communication technologies in fruit and vegetable supply chain: a systematic literature review

Yuvaraj

Basu

Abdulrahman

et al. 2023

IMDS

View full text Add to dashboard Cite

PurposeInformation and communication technology (ICT) implementation has demonstrated usefulness in supply chain coordination and efficiency optimization in various industries and sectors. This study investigates the extent of ICT deployment in fruits and vegetable supply chains (FVSC) from “farm-to-fork” to ensure food security.Design/methodology/approachThis paper employs a systematic literature review (SLR) methodology and identified a total of 99 journal articles ranging from 2001 to April 2023 for analysis. The reviewed articles have been classified based on the framework proposed from the perspective of food security. Bibliometric and content analysis is carried out with the final list of articles to extract useful insights.FindingsThe findings reveal that ICT implementation in FVSC is a relatively new research area; researchers have started investigating several aspects of ICT in FVSC through varied research methodologies. Experimental research aimed at addressing food safety and condition monitoring of fruits and vegetables (FV) has started to gain traction while theory building is yet to gain traction in the literature reviewed. Findings indicate further research is required on technologies like blockchain (BCT), artificial intelligence (AI) and machine learning (ML), especially on key objectives such as food security, and the triple-bottom-line approach of sustainability. It also indicates that implementing relevant ICTs in FVSC can help delay, if not avert, the food crisis predicted by Malthusian theory.Research limitations/implicationsThis study used only well-established databases to ensure quality of the studies examined. There is a possibility of missing out on articles from other sources not considered. As a result, future SLR studies may employ additional databases, such as Springer Link, Taylor and Francis, Emerald Insight and Google Scholar. Other methodologies such as expert interviews and extra empirical methodologies may also be employed to give a more balanced picture and insights into ICTs implementation in FVSC.Practical implicationsThis study offers a summative detail of the status of ICT implementation in FVSC and can serve as a reference guide for stakeholders in developing strategies for efficient FVSC management. This research work highlights the impact of ICT implementation in FVSC on the four pillars of food security which include improved availability, accessibility, utilization and stability.Originality/valueThis study focuses on ICT implementation for food security in FVSC. The SLR highlights the gaps and proffers potential solutions that enhance global efforts on food security through ICT-enabled reduction in food waste and food loss in FVSC.

show abstract

Section: Results and Analysismentioning

confidence: 99%

Implementation of information and communication technologies in fruit and vegetable supply chain: a systematic literature review

Yuvaraj

Basu

Abdulrahman

et al. 2023

IMDS

View full text Add to dashboard Cite

show abstract

“…On the other hand, deep learning models require complex architectures with many parameters requiring significant computational resources and time-consuming training procedures. Furthermore, optimization tasks are quite complicated using deep learning [56].…”

Section: E Classical Machine Learning For Fruit Damage Detectionmentioning

confidence: 99%

A Review on Automated Detection and Assessment of Fruit Damage Using Machine Learning

Safari,

Nakatumba-Nabende,

Nakasi

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Automation improves the quality of fruits through quick and accurate detection of pest and disease infections thus contributing to the country's economic growth and productivity. Although humans can identify the fruit damage caused by pests and diseases, methods used are inconsistent, time-consuming, and variable. The surface features of fruits typically observed by consumers who seek their health benefits, affect their market value. The issue of pest and disease infections further deteriorates fruits' quality, becoming a mounting stressor on farmers as they affect the potential income that could have been realised from production, processing and export. This article reviews various studies on detecting and classifying damages in fruits. Specifically, we review articles where state-of-the-art approaches under segmentation, image processing, machine learning, and deep learning have proved effective in developing automated systems that address hurdles associated with manual methods of assessing damage using visual experiences. This survey reviews 32 Journal and Conference articles spanning 13 years obtained electronically through Google Scholar, Scopus, IEEE, ScienceDirect, and general internet searches. This survey further presents a detailed discussion of related studies done in the past while emphasizing their strengths and limitations and presenting future research directions. It also reveals that much as the use of automated detection and classification of fruit damage has yielded promising results in the horticulture industry, more research is still needed with systems required to fully automate the detection and classification processes, especially those that are mobile phone-based towards addressing occlusion challenges.INDEX TERMS fruit damage detection, classification, deep learning, image analysis and segmentation.

show abstract

“…During postharvest, there could be considerable losses of fruits due to poor handling, grading and distribution practices (Bantayehu et al, 2017;Singh et al, 2022;Knott et al, 2023). Machine learning and deep learning algorithms are currently being used for fruit transport and storage as well as fruit sorting and distribution after harvest by analyzing images and sensor and historical data (Knott et al, 2023). For example, machine learning algorithms can be used to detect defects, diseases and damage to fruit, as well as to estimate fruit quality, maturity and shelf life.…”

Section: Group Search Stringmentioning

confidence: 99%

Assessing the impact of emerging technologies on sustainable fruit production: A systematic review of the literature

Pardo-Pardo,

Cuervo-Bejarano

2023

Agron. Colomb.

View full text Add to dashboard Cite

Agriculture 4.0 refers to innovations in technological tools used in agriculture to achieve different objectives, such as adapting the supply chain to avoid waste, increasing productivity and collecting mass data through ICT (Information and Communication Technologies) to meet the growing food demand of the population. The objective of this study is to conduct a systematic literature review to evaluate the impact of emerging technologies on sustainable fruit production. Initially, a bibliographic search was conducted on the technologies currently implemented in agriculture; the Bibliometrix library of the R Studio software was used, and then an analysis of relevant scientific publications published in the last ten years was carried out through the VOSviewer® software, which allowed the construction and visualization of bibliometric networks. The results show Europe and China as the leading regions in technological development, while developing countries face economic and research limitations; in Colombia, the use of Agriculture 4.0 is focused on the implementation of satellite images for monitoring agro-climatic conditions. In summary, Agriculture 4.0 aims to achieve economic, social, and environmental sustainability in the agri-food sector through data-generating technologies to improve production, reduce costs, and ensure food safety and quality. However, there is a technology gap between developed and developing countries that affects the adoption of these innovations. More support is therefore needed from governments, academia, and the private sector to drive innovation, training, and adoption of these technologies, which can contribute to the economic, social, and environmental development of the country.

show abstract

Facilitated machine learning for image-based fruit quality assessment

Cited by 21 publications

References 18 publications

Implementation of information and communication technologies in fruit and vegetable supply chain: a systematic literature review

Implementation of information and communication technologies in fruit and vegetable supply chain: a systematic literature review

A Review on Automated Detection and Assessment of Fruit Damage Using Machine Learning

Assessing the impact of emerging technologies on sustainable fruit production: A systematic review of the literature

Contact Info

Product

Resources

About