Development of a Smartphone-Based Remote Sensing System for Precision Crop Monitoring

Thesis Abstract

In the quest to enhance agricultural productivity and resource management, effective crop monitoring remains a critical component, yet existing remote sensing systems often face limitations related to cost, accessibility, and real-time data acquisition, particularly in resource-constrained settings. This study aims to develop a scalable, cost-effective, and user-friendly smartphone-based remote sensing system tailored for precision crop monitoring, thereby bridging the gap between advanced remote sensing technologies and practical farm-level application. The specific objectives include designing and implementing a mobile application capable of capturing multispectral imagery, evaluating the system’s accuracy in estimating crop health indices, and assessing its usability and acceptance among smallholder farmers. This investigation adopts a mixed-method research design, integrating quantitative analysis of sensor accuracy with qualitative assessments of user experience. The study's population comprises 150 smallholder farmers across two agricultural districts with varying crop types, selected through stratified random sampling to ensure diversity in landholding size, crop varieties, and technology familiarity. Data collection involves the deployment of 150 custom-developed smartphone sensor kits and a structured questionnaire, supplemented by semi-structured interviews to gauge user perceptions and contextual challenges. Quantitative data analyses include regression analysis to examine correlations between sensor-derived indices and conventional measurements, while thematic analysis is employed on qualitative data to explore user acceptance and perceived system benefits and limitations. The study also applies the Technology Acceptance Model (TAM) to interpret factors influencing adoption. It is anticipated that the developed system will demonstrate high correlation coefficients (above 0.85) with satellite-based remote sensing data in estimating crop vigor indices such as NDVI, with an overall accuracy of 87%. The findings are expected to show that the smartphone sensing system is feasible and accessible for smallholder farmers, with positive indications of ease of use, perceived usefulness, and willingness to adopt. These results contribute novel insights into mobile-based remote sensing solutions, emphasizing affordability and operational simplicity, especially in developing country contexts. Additionally, the study advances the theoretical understanding of technology acceptance in rural farming communities by contextualizing TAM within the framework of precision agriculture. The primary conclusion underscores the potential of smartphone-integrated remote sensing as a practical tool for real-time crop monitoring, supporting timely decision-making and resource optimization. Based on these findings, recommendations include scaling up the system through integration with existing agricultural extension services, further validation in diverse agricultural landscapes, and the incorporation of machine learning algorithms to enhance predictive capabilities. Ultimately, this research demonstrates that leveraging ubiquitous mobile technology can democratize remote sensing and foster sustainable intensification in crop production systems, thereby making a valuable contribution to both academic knowledge and practical agricultural development.