Current issue No 4/2025, December | Agris on-line Papers in Economics and Informatics

A Highly Effective Deep Learning Tool for Identifying Plant Leaves

Adel Abdelhadi Ouahab Kadri DOI: 10.7160/aol.2025.170401
No 4/2025, December
pp. 3 - 9

ABSTRACT

This work addresses pattern recognition in the agronomic domain, with a particular emphasis on identifying plant leaves using an adaptive neural network technique. We introduce a tool designed for two primary groups: botany researchers and a broader range of scientists applying it to plant identification and classification. We delve into the capabilities of Deep Learning, focusing on generalization abilities that enable accurate predictions on unseen data, which is essential for handling the variation in leaf shapes, sizes, and structures across species. The implementation details of these neural networks are described, including data preprocessing, network architecture design, training strategies, and evaluation techniques to ensure robustness and reliability in real-world applications.

KEYWORDS

Pattern recognition, plant leaves, deep learning, classification, analysis, image processing, neural networks.

VIEW MORE PDF (.pdf, 1022.18 KB)

Factors Influencing Consumer Preference Towards Horticulture Geographical Indications (GIs): A Case of Udupi Brinjal from South India

Vikram Baliga Santhosha Shetty G Krithika Rao Mayuri Ren Trevor Dias DOI: 10.7160/aol.2025.170402
No 4/2025, December
pp. 11 - 26

ABSTRACT

Consumers' attention towards fruits and vegetables in recent times has shifted to regionally grown geographical indications (GI) due to the quality and origin of these products. This research work aims to explore the factors influencing consumer preference towards Udupi Brinjal, a horticulture GI grown in Udupi District of South India. The present study has used a mixed-method approach to gather and analyze data collected from local consumers of Udupi District. The qualitative study design involved a survey of key informants in the local region. Subsequently, data collected from consumers through a structured questionnaire were analyzed using factor analysis and regression techniques. Results from data analysis revealed that quality factors show greater importance in predicting consumer preference, followed by sensory attributes and health-related aspects. The results will help formulate marketing strategies for horticulture GIs. Agri-business marketers and State-owned agriculture promotion agencies can adopt these strategies to promote GIs and gain consumer acceptance. The results and discussions of this research work are consistent with Sustainable Development Goals (Goals 2 & 12) and contribute to sustainable agriculture.

KEYWORDS

Consumer preference, Geographical Indications (GIs), sustainable agriculture, marketing strategies, horticulture marketing, territorial specialties, region-of-origin products.

VIEW MORE PDF (.pdf, 541 KB)

Leveraging Deep Learning for Early Detection and Diagnosis of Wheat Diseases: Challenges and Innovations

Chemesse Ennehar Bencheriet Hala Hamouchi Mohamed Islem Hadri DOI: 10.7160/aol.2025.170403
No 4/2025, December
pp. 27 - 36

ABSTRACT

This research introduces a deep learning system for the early identification and categorization of wheat illnesses, with the objective of optimizing crop health and promoting agricultural sustainability. Results in up to high classification accuracy for brown rust, yellow rust, leaf rust, and septoria. The combination of artificial intelligence (AI) with image processing methodologies such as rescaling and augmentation allows the system to accurately classify wheat crops that are well or unhealthy. The presented system is of great interest for precision agriculture, providing an affordable means to reduce the application of pesticides and encourage sustainable agricultural practices. Ongoing research involves linking this diagnostic platform with drone technology to facilitate on-demand, point-by-point disease surveillance and monitoring across large areas, further extending the platform’s applicability in field applications for food securit.

KEYWORDS

Wheat disease detection, deep learning, convolutional neural networks (CNNs).

VIEW MORE PDF (.pdf, 1.58 MB)

Impact of Rural Out-Migration on Crop Productivity of Migrant-Sending Rural Households in Oromia Region of Ethiopia

Fassil Eshetu Semeneh Bessie Lamessa T. Abdisa Abdulaziz Dawud Fekadu Abdissa DOI: 10.7160/aol.2025.170404
No 4/2025, December
pp. 37 - 56

ABSTRACT

This study quantified the impact of rural out-migration on crop productivity using the multinomial endogenous switching model as an analytical model in the Oromia region of Ethiopia. Cross-sectional data were gathered from a random sample of 384 rural households. The descriptive analysis revealed that the rate of rural-rural migration in Ethiopia decreased from 55.8 to 24.6% while the rate of rural-urban migration increased from 28.7 to 33.8 % between 1984 and 2021. The proportion of migrants in the total urban population increased from 17.2 to 49.2% in the Oromia region between 1999 and 2021. The regression results found that land size, use of irrigation, tropical livestock unit, dependency ratio, and education level of household head decrease the likelihood of participating in migration, whereas family size, number of plots, being female-headed households, and age of household head increase the probability of participating in migration. The participation in rural-urban and international migration increases the productivity of wheat producers by 341.28 and 707.21kilograms, respectively. Similarly, the participation in rural-urban and international migration increases the productivity of teff producers by 502.05 and 257.04 kilograms, respectively. This finding also supports the credit and risk hypotheses of the new economics labour migration theory. Enhancing access to finance or credit markets, agricultural land, and enhanced technology for youth in migrant-sending rural communities can leverage the gains from rural out-migration. Provision of pre-migration training, rural non-farm employment, awareness creation, promotion of safe migration, and better rural public services would capitalize the net benefit from out-migration.

KEYWORDS

Migration, new economics labor migration, productivity, switching model, Oromia.

VIEW MORE PDF (.pdf, 848.28 KB)

Farmer Involvement in Irrigation Agriculture: Evidence from the Anambra-Imo River Basin Irrigation Scheme, Nigeria

Ubon Essien Okwudili Ibeagwa Igwe Ukoha Sylvanus Ben Onyekachukwu Ejike DOI: 10.7160/aol.2025.170405
No 4/2025, December
pp. 57 - 66

ABSTRACT

This study was conducted in the Anambra catchment of the Anambra-Imo River Basin Development Authority(AIRBDA), Nigeria. The aim was to analyse the involvement of farmers in irrigation agriculture as a key component of public agricultural project performance. A multi-stage sampling procedure was adopted in selecting ninety(90) farmers from the catchment of the AIRBDA. Descriptive statistics provided initial insight into operational and structural characteristics, while relevant visualizations were produced using Python and Excel. The Logit estimate identified factors influencing farmers’ involvement in the irrigation schemes, thereby offering empirical evidence relevant for project appraisal and management. Results showed that 15.6% of farmers reported non-participation, while about 84.4% were active participants in the scheme. The estimated model reported a Wald chi² of 39.65 and a log pseudolikelihood of –281.37084. Farm experience, household size, major occupation, farm income, and membership in the Water Users Association (WUA) significantly influenced farmers’ involvement in the irrigation scheme. It recommends strengthening Participatory Irrigation Management (PIM) systems, whereby farmers manage routine water allocation, while the River Basin management provides technical oversight, with a member of the Water Users Association as a part of its team.

KEYWORDS

Irrigation scheme, project, involvement, farmers, logit model, Anambra-Imo-River-Basin

VIEW MORE PDF (.pdf, 647.94 KB)

Relationship Between Rural Poverty and Agricultural Diversification at a Local Scale in Colombia: An Approach through Spatial Effects

Alejandro Mojica Godoy DOI: 10.7160/aol.2025.170406
No 4/2025, December
pp. 67 - 70

ABSTRACT

This paper examines the relationship between local agricultural diversification and rural poverty in Colombia. The evidence presented suggests that municipal-scale agricultural diversification is associated with higher levels of rural poverty. The primary mechanism driving this relationship is the loss of external economies of scale in agricultural production. However, when analyzing the spatial effects (autocorrelation and spatial heterogeneity), it was found that this relationship varies across the country. Neighboring municipalities have opposite spillover effects that offset the positive effect. In addition, the presence of small rural producers, greater provision of public goods, the existence of industrial crops, and lower persistence of acts of violence were all found to be associated with lower levels of rural poverty. This study joins the literature on the economic and social effects of agricultural diversification, particularly in the context of its promotion as a means of adapting to and mitigating the effects of climate change

KEYWORDS

Agricultural diversification, poverty, spatial econometrics, spillover effects, regional economics

VIEW MORE PDF (.pdf, 2.2 MB)

Data-Driven Optimisation of Irrigation Dose Using Machine-Learning Ensembles for Sustainable European Agriculture

Zuzana Palková Miroslav Žitňák Jan Valíček Marta Harničárová Miroslav Holý Daniel Levák Hakan Tozan Karol Görči DOI: 10.7160/aol.2025.170407
No 4/2025, December
pp. 81 - 99

ABSTRACT

This study focuses on predicting irrigation doses using digital technologies and statistical modelling to enhance water resource management in agriculture. Conducted as part of the CODECS project in the semi-arid Nitra region of Slovakia, this study aimed to evaluate the effectiveness of various irrigation systems and to develop predictive models for optimal irrigation doses. The methodology integrates environmental sensor data, agronomic models, and machine learning techniques, utilizing IoT sensors alongside Valley and Irriga control software. A significant challenge was the incompatibility of heterogeneous data from different sources, leading to the creation of a unified method-ology for data collection, validation, and analysis. Analytical tools, such as ex-ploratory data analysis, correlation techniques, and regression models, were employed to identify key factors affecting irrigation efficiency, including precipitation, temperature, soil moisture, and energy consumption. The findings aim to inform sustainable irrigation strategies that reduce water usage, enhance crop productivity, and safeguard soil resources under changing climatic con-ditions.

KEYWORDS

Artificial irrigation, digital agriculture, machine learning, data, statistical modelling, smart farming

VIEW MORE PDF (.pdf, 1.4 MB)

Enhancing Market Access for Smallholder Farmers in Indonesia: The Role of Managerial Capacity and Member Motivation in Collective Action within Farmer Groups

Rokhman Permadi Lili Winarti DOI: 10.7160/aol.2025.170408
No 4/2025, December
pp. 101 - 115

ABSTRACT

competitive markets. The objective is to investigate the impact of managerial capacity and member motivation on collective action and market access among smallholder farmers in Indonesia. A survey was conducted with 249 kepok banana farmers belonging to farmer groups in Seruyan Regency, Central Kalimantan, Indonesia. Data were collected using a structured questionnaire that included demographic information and perceptions of managerial capacity, motivation, collective action, and market access. Partial Least Squares Structural Equation Modeling (PLS-SEM) was employed to assess the relationships among the constructs. The results indicate that managerial capacity significantly enhances the role of farmer groups (β = 0.494, p < 0.001), while member motivation does not show a significant effect (β = 0.076, p = 0.290). The role of farmer groups significantly influences both collective action (β = 0.616, p < 0.001) and market access (β = 0.240, p < 0.001). Furthermore, collective action has a significant positive effect on market access (β = 0.479, p < 0.001). These findings underscore the critical role of farmer groups in organizing collective strategies to enhance market access. Managerial capacity is pivotal for successful collective action. Policymakers should strengthen farmer group institutions to foster collective action, reduce market barriers, and achieve sustainable agricultural growth.

KEYWORDS

Agricultural development, economic resilience, institutional capacity, PLS-SEM

VIEW MORE PDF (.pdf, 2.63 MB)

Development of a Supply Chain Management Platform for Rubberwood Biomass in Southern Thailand

Sirirat Pungchompoo Nikorn Sirivongpaisal Sirirat Suwatcharachaitiwong Dollaya Buakum DOI: 10.7160/aol.2025.170409
No 4/2025, December
pp. 117 - 135

ABSTRACT

This research aims to manage biomass raw materials in line with industrial needs by developing a platform that links stakeholders in the rubberwood biomass supply chain in southern Thailand. Geographic Information System (GIS) technology was applied to build a database and estimate rubber plantation areas. The trees were grouped by age into three categories: 14–20, 21–27, and over 27 years. The platform also provides information on garden and factory locations, including sawmills, rubberwood processing plants, biomass production plants, and biomass power plants in 14 southern provinces. The system, available on Android and iOS, supports users in making decisions about transportation costs such as distance, time, and fuel. Results from the technology transfer show that the platform is practical, matches user requirements, and is used effectively. The average user satisfaction scores were 4.538 for function and 4.504 for overall use, reflecting the platform’s usefulness and acceptance among stakeholders.

KEYWORDS

Rubberwood supply chain, rubberwood biomass, rubberwood biomass platform.

VIEW MORE PDF (.pdf, 2.07 MB)

Comparative Advantages and Specialization Dynamics in Agri-food Trade of Argentina, Paraguay and Uruguay

Ivo Zdráhal Martin Hrabálek Petr Kadlec Oldřich Krpec DOI: 10.7160/aol.2025.170410
No 4/2025, December
pp. 137 - 155

ABSTRACT

The article interrogates the shape, dynamic, and fragility of revealed comparative advantages of 46 agri-food products traded by Argentina, Paraguay, and Uruguay in the period 1995-2020 using normalized revealed comparative index and summary statistics, stochastic kernels, Galtonian regression, Markov chains, and Kaplan-Meier survival analysis. The analysis reveals the agri-food flagship products and the agri-food trade of these countries has formed mainly around these flagship products. The results support the argument that changes in distribution of comparative advantages in agri-food trade underwent an increase in specialization in these countries, especially in the period from the beginning of millennia until about period slightly after the Great Recession. The results also indicate slight convergence in the change in agri-food comparative advantages in these three countries, as well as the increased complexity of agri-food comparative advantages in Argentina and Paraguay at the end of the period under scrutiny. Despite these variations and differences among countries under scrutiny the distribution of comparative advantages remains stable and persistent. Given this evidence, we conclude that these countries will continue to develop their agriculture-led growth economic model and these flagship products will play an important role in the overall agri-food export structures of these countries in the future.

KEYWORDS

Liberalization, Markov chain model, regression analysis, revealed comparative advantage, RCA, agri-food trade.

VIEW MORE PDF (.pdf, 861.03 KB)