Recent Developments of Artificial Intelligence for Banana: Application Areas, Learning Algorithms, and Future Challenges

Abstract

Bananas are the world's most traded fruits. Several analytical models using artificial intelligence (AI) have been developed to resolve challenges facing the banana supply chain. The number of publications in this field has steadily increased each year. However, a literature review regarding the trends of recent AI developments is not available. Thus, this study reviews the current scenario of scientific research involving AI in the stages of the banana supply chain (pre-harvest, harvest, post-harvest, processing and retail). This review covers literature published between 2015 and 2020 from online databases. Fiftytwo relevant studies were retrieved from 23 countries. Consequently, we propose an AIperformance framework based on real applications implemented for bananas: the application domain, learning algorithms, performance metrics, and reported impacts. This paper discovers 11 AI-application areas for bananas, such as ripeness, leaf diseases, quality grading, crop type, crop yield, and soil control. Moreover, this review summarizes the main functionality of learning algorithms found in the literature (ANN, CNN, SVM, and K-NN). Finally, the future challenges are discussed. This comprehensive review will help researchers understand AI applications in the banana sector and analyze the knowledge gap for future studies.

Exploring computer vision, machine learning, and robotics applications in banana grading: A review

“…When capturing images of bananas, only one direction is tested, but more angles are required to cover up during the image acquisition, which also entails a challenge in this direction [36]. In machine learning, issues related to computational resources for training and testing algorithms, and userfriendly predictive models for farmers and other end-users, present logistical challenges [37]. Additionally, robotics for the gentle handling of bananas presents challenges, where a requirement for advanced mechanization systems to ensure the quality of banana hands (clusters or bunches of bananas) and prolong the transportation time is needed as banana hands are susceptible to mechanical damage [38].…”

Detecção de sigatoka amarela em folhas de banana com redes neurais convolucionais (CNN) e aprendizagem por transferência profunda (DTL)

“…A seção de trabalhos relacionados discute estudos anteriores e pesquisas relevantes diretamente relacionadas ao tema deste trabalho. No contexto abordado, [7] analisa as últimas tendências e desafios relacionados à aplicação da inteligência artificial na produção de bananas. A utilização dessas técnicas pode trazer melhorias significativas para a qualidade e produtividade da produção de bananas, além de permitir uma detecção mais precisa de doenças e pragas nas plantações.…”

“…Essas desvantagens do controle químico motivam a busca por alternativas mais eficazes e sustentáveis. Nesse contexto, o desenvolvimento de métodos alternativos de detecção automatizada, com o uso de modelos de redes neurais convolucionais (Convolutional Neural Networks -CNNs), pode prover uma abordagem promissora para a detecção rápida e precisa da sigatoka amarela em folhas de bananeira [7]. Esse modelo pode auxiliar os agricultores no monitoramento das plantações, permitindo a detecção precoce de áreas afetadas pela doença, implementando medidas de controle direcionadas, resultando em um maior rendimento da produção e redução das perdas econômicas.…”