Applications of sensor-based technologies in vegetable crops: a narrative review

Sensor-based technologies are increasingly used in vegetable crops to characterize plant traits, monitor crop health, and assess quality throughout the production cycle. Conventional approaches for measuring plant traits and crop performance remain constrained by low throughput, high labor requirements, and subjective variability. Recent advances in sensing and imaging technologies have enabled precise, non-destructive, high-throughput, and spatially and temporally resolved assessment of vegetable crops. This narrative review provides a detailed synthesis of sensor-based technologies and their applications in vegetable crop phenotyping, with emphasis on morphological, physiological, biochemical, and physical trait phenotyping, as well as preharvest stress and health monitoring and postharvest assessment. Major imaging and sensing modalities, including RGB, multispectral and hyperspectral imaging, thermal and fluorescence imaging, LiDAR, VIS-NIR spectroscopy, and wearable and flexible sensors, are discussed in terms of their principles, capabilities, applications, and limitations. The integration of these sensors with proximal platforms (handheld and ground-based systems) and remote platforms such as unmanned aerial vehicles enables multi-scale data acquisition and rapid monitoring of crop performance. Data processing and analytical approaches, including image and spectral analysis, machine learning, and deep learning, are discussed as important components for extracting meaningful information from sensor-derived data. Key challenges, such as sensor calibration, environmental variability, data integration, and model transferability, are also discussed. Overall, the review highlights the expanding role of sensor-based technologies in vegetable crop phenotyping, crop monitoring, and postharvest assessment and provides a decision matrix to guide the selection of sensing strategies according to crop group, target trait, and spatial scale, recognizing that suitable combinations depend on the specific applications and conditions.