The Hybrid Deep Learning Approach for Cotton Leaf Disease Detection and Management Using Fine-Tuned VGG16 and Inception v3 Models

Ravikumar ch — Wed, 06 May 2026 00:00:00 +0800

Agricultural losses cause significant barriers to the practice of sustainable farming, and the only way to prevent them is through earlier interventions. This paper introduces a novel technique of Hybrid Deep Learning Method (HDLM) to augment the classification accuracy of the diseases by using a voting system at the output layers on the tuned VGG16, Inception v3, ResNet-50, and MobileNet-v2 models. The model was trained and tested on a structured dataset of 3,000 images, which upon manual curation, were categorized into Aphids, Armyworm, Bacterial Blight, Powdery Mildew, Target Spot, and Healthy Leaves. Out of the 3,000, 2,400 were labeled for the model to learn on and 600 were held for validation. In order to improve the model's external predictive ability, strategies for data augmentation were used. The proposed method of HDLM achieved an accuracy of 98.56% which is a significant improvement over the benchmark models, namely AlexNet, DenseNet-121, ResNet-50, LeNet-5, and a 7-layer CNN, which report an accuracy between 90 and 95 percent. The model also features a disease management recommendation system which provides users with relevant strategies proven to aid farmers and agricultural constituents. The research demonstrates the value addition offered by ensemble deep learning to the field of plant disease detection. In the future, the focus is to broaden the model’s adaptability and effectiveness across different agricultural settings by expanding the dataset and utilizing more sophisticated augmentation methods.

Reframing Spatial Housing Models: A Lateral Thinking Approach to Real Estate Valuation in Taichung, Taiwan

Jyhjeng Deng, Hsun-Yu Chan, Yu-Chen Lai, Yu-Xin Huang — Wed, 06 May 2026 00:00:00 +0800

This study investigates the impact of geographic features on housing prices, based on 229 transactions recorded between September 2023 and September 2024 in the North District of Taichung City, Taiwan, using data from the Ministry of the Interior’s real estate platform. Building upon an existing regression model that includes structural variables such as the lot size, number of rooms, age, number of floors, availability of an elevator, and the presence of a garage, we introduce six additional geographic variables that reflect the proximity to a metro station, hospital, museum, central business district (CBD), funeral home, and clothing outlets. The revised model demonstrates improved explanatory power, with the R² increasing from 0.756 to 0.791. Among the newly included variables, proximity to the CBD and the museum exhibit the strongest positive effects on housing prices, contributing increases of 988.3381 and 616.4406 Wons, respectively. These effects are both statistically significant. In contrast, proximity to a metro station has a significant negative impact of –722.7214 Wons, a finding that is attributed to a boundary constraint, since only sales south of the station fall within the district, creating a data bias. Using the escape technique of lateral thinking these results also provide practical guidance for buyers by identifying cost-effective areas located just outside the 500 m proximity zones of high-value amenities. One such example is located near the intersection of Zhongming Road and Section 1, Zhongqing Road, with price savings of at least 736 Wons. A residual analysis reveals clustering in the Yizhong Business District, suggesting that other unobserved value factors may warrant future investigation.

International Journal of Systematic Innovation

The Hybrid Deep Learning Approach for Cotton Leaf Disease Detection and Management Using Fine-Tuned VGG16 and Inception v3 Models

Reframing Spatial Housing Models: A Lateral Thinking Approach to Real Estate Valuation in Taichung, Taiwan