Multimodal Crop Phenotyping from Unmanned Platforms: Detecting Open Cotton Bolls with LiDAR and RGB

Abstract

Precision Agriculture Sensing is advancing through efforts to align performance with evidence quality, resource limits, and transfer across settings. This comparative analysis considers combining geometry and appearance for repeatable field-scale crop measurement. The corpus joins 1 focal paper with 12 independently retrieved publications confirmed at bibliographic registration or publisher level. The analysis is organized around sensor registration, point-cloud structure, appearance cues, occlusion, and agronomic validation. The synthesis resists treating published metrics as automatically comparable, the review compares problem formulation, design assumptions, and transfer boundary. Across the literature, the central lesson is that advances in precision agriculture sensing become credible when representation, objective, and evaluation protocol are evaluated together and when uncertainty about distribution shift is reported explicitly. The framework consequently connects method selection to application risk, identifies recurring threats to external validity, and proposes a research agenda centered on clear comparators, adversarial conditions, and inspectable evidence.