Journal of Business, Social and Technology
Vol. 7 No. 4 (2026): Journal of Business, Social and Technology

Diminishing Returns of Algorithmic Complexity: A Cost-Benefit Evaluation of UAV Image Enhancement for Precision Weed Detection in Rice Farming

Huda, Muhamad Syaiful (Unknown)
Setiyono, Budi (Unknown)



Article Info

Publish Date
13 Oct 2026

Abstract

Background: Unmanned aerial vehicle (UAV) imagery is increasingly used for weed detection in rice fields, but limited spatial resolution can constrain segmentation performance. Image enhancement may improve detection; however, it remains uncertain whether computationally complex, deep learning–based enhancement provides task-level gains beyond those achieved through classical interpolation. Objective: This study evaluates whether increasing enhancement complexity yields additional weed-detection benefits across four image conditions. Methods: Using 148 UAV images from the WeedyRice-RGBMS-DB dataset, intersection over union (IoU) was measured under four image conditions: no enhancement, nearest-neighbor interpolation, bicubic interpolation, and an image-restoration enhancement (IRE) method. The U-Net model with an EfficientNet-B5 backbone was held constant across all conditions. Paired differences were evaluated using the Wilcoxon signed-rank test with Holm–Bonferroni correction, and effect sizes were quantified using Cohen’s d. Results: All enhancement conditions outperformed the no-enhancement condition (adjusted p < 0.001). Nearest-neighbor interpolation performed worse than both bicubic interpolation and IRE (adjusted p < 0.02), whereas bicubic interpolation and IRE did not differ significantly (adjusted p = 0.937; d = 0.019), despite IRE achieving higher peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) values. Conclusion: For this dataset and detector, the task-level benefits of image enhancement plateaued at bicubic interpolation. This finding supports a cautious technology-selection principle: more complex enhancement methods should be justified by measurable downstream gains, while claims of economic superiority require direct measurement of computing, hardware, licensing, and implementation costs.

Copyrights © 2026






Journal Info

Abbrev

jbt

Publisher

Subject

Economics, Econometrics & Finance Industrial & Manufacturing Engineering Social Sciences

Description

This journal publishes research articles covering all aspects of information technology, information systems, agricultural technology, computer social and political sciences, and economics that belong to the business, social, and technological ...