Computer Vision-Based Assessment of Environmental Stress and Fungal Pathologies in Aloe vera within the Churu Agro-Climatic Zone

The hyper-arid and extreme continental climate of the Churu district in western Rajasthan presents severe environmental challenges for succulent medicinal agriculture. Characterized by shifting sand dunes, low precipitation, high solar irradiance, and extreme thermal amplitude ranging from sub-zero winter nights to summer temperatures exceeding 〖48〗^∘ "C" the Churu microclimate subject’s Aloe vera (Ghritkumari) to chronic abiotic stress. These environmental pressures frequently overlap with or predispose the crop to biotic fungal infections, most notably leaf spot (Alternaria alternata) and basal collar rot. Traditional visual scouting methods are labor-intensive, highly subjective, and frequently fail to distinguish between non-pathological physiological adaptations (such as UV-induced anthocyanin reddening and tip desiccation) and early-stage fungal lesions. This paper introduces a specialized computer vision framework tailored explicitly for field-captured imagery of Aloe vera under unconstrained Churu environmental conditions. By integrating deep convolutional neural network (CNN) feature backbones (ResNet-50) with mathematical spatial descriptors—specifically Box-Counting Fractal Dimension (D), Grey-Level Co-occurrence Matrix (GLCM) texture attributes, and CIELAB (L^* a^* b*) color space transformations—the model successfully decouples abiotic stress manifestations from necrotic fungal boundaries. Evaluated on an extensive dataset of 4000 field-verified images collected across diverse rural plots in Churu, the proposed hybrid ensemble framework achieved an overall classification accuracy of 95.8%, outperforming standard baseline vision models by over 5.4%. This research provides a robust, automated diagnostic pipeline for precision agriculture and phyto medicinal conservation in desert ecosystems.

Authors

Institutions

Publication Details

Journal
International Research Journal of Computer Science
Published
2026-11-17
DOI
https://doi.org/10.26562/irjcs.2026.v1311.01
Primary Topic
Smart Agriculture and AI
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Computer Vision-Based Assessment of Environmental Stress and Fungal Pathologies in Aloe vera within the Churu Agro-Climatic Zone

Prof.(Dr).Harvir Singh, Mitesh Kumar
International Research Journal of Computer Science
Smart Agriculture and AI
article

Computer Vision-Based Assessment of Environmental Stress and Fungal Pathologies in Aloe vera within the Churu Agro-Climatic Zone

Prof.(Dr).Harvir Singh, Mitesh Kumar
article en

Abstract

The hyper-arid and extreme continental climate of the Churu district in western Rajasthan presents severe environmental challenges for succulent medicinal agriculture. Characterized by shifting sand dunes, low precipitation, high solar irradiance, and extreme thermal amplitude ranging from sub-zero winter nights to summer temperatures exceeding 〖48〗^∘ "C" the Churu microclimate subject’s Aloe vera (Ghritkumari) to chronic abiotic stress. These environmental pressures frequently overlap with or predispose the crop to biotic fungal infections, most notably leaf spot (Alternaria alternata) and basal collar rot. Traditional visual scouting methods are labor-intensive, highly subjective, and frequently fail to distinguish between non-pathological physiological adaptations (such as UV-induced anthocyanin reddening and tip desiccation) and early-stage fungal lesions. This paper introduces a specialized computer vision framework tailored explicitly for field-captured imagery of Aloe vera under unconstrained Churu environmental conditions. By integrating deep convolutional neural network (CNN) feature backbones (ResNet-50) with mathematical spatial descriptors—specifically Box-Counting Fractal Dimension (D), Grey-Level Co-occurrence Matrix (GLCM) texture attributes, and CIELAB (L^* a^* b*) color space transformations—the model successfully decouples abiotic stress manifestations from necrotic fungal boundaries. Evaluated on an extensive dataset of 4000 field-verified images collected across diverse rural plots in Churu, the proposed hybrid ensemble framework achieved an overall classification accuracy of 95.8%, outperforming standard baseline vision models by over 5.4%. This research provides a robust, automated diagnostic pipeline for precision agriculture and phyto medicinal conservation in desert ecosystems.

International Research Journal of Computer ScienceVol. 13(11)
Dr. K.N.Modi University (IN)
Smart Agriculture and AI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon