Hyperspectral imaging (HSI) is a powerful technique capable of obtaining both spatial and spectral information from a target by combining conventional machine vision and point spectroscopy methods. During the past two decades, the HSI technologies have found numerous successful applications for food and agriculture. Owing to growth in data, advances in machine learning algorithms and software, and increased computing power, artificial intelligence (AI) is currently driving a new wave of innovation that is poised to revolutionize many industries. Sensing technologies in various disciplines are becoming one of primary frontiers for the AI-driven developments and transformations. By leveraging the combination of the HSI and learning-based AI techniques, it is promising to accomplish novel sensing applications with measurement capabilities and accuracies that were not possible in the past.
This Research Topic intends to collect papers that highlight recent advances in research and development of AI-based hyperspectral imaging methodologies and their applications in the food and agricultural areas. Recently, miniature imaging spectrographs/spectrometers, all-in-one small hyperspectral cameras, and snapshot hyperspectral imagers are commercially available, which make it possible to develop compact and portable imaging systems and devices that can be used in confined space (e.g., overhead imaging in plant growth chambers) and field applications (e.g., crop health/disease sensing using drones or ground platforms). Learning-based data analysis methods, along with high-performance computing, open a new avenue to process hyperspectral images for classification and regression purposes. Integration of AI capabilities through machine learning into hyperspectral imaging systems is becoming a new axis to develop novel smart sensing devices and instruments. Articles presenting the development of compact and portable spectral imaging systems, multimodal hyperspectral imaging and data fusion methods, advanced machine learning and deep learning algorithms for hyperspectral data analysis, embedded AI techniques for real-time inspections, as well as AI-based hyperspectral applications for food, precision agriculture, and high-throughput phenotyping are of particular interest.
This Research Topic welcomes Original Research, Review, Mini Review, and Perspective manuscripts on themes including but not limited to:
• Compact and portable hyperspectral and multispectral systems and devices for smart sensing
• Single and multimodal hyperspectral imaging and data fusion (e.g., reflectance, fluorescence, Raman, and terahertz)
• Machine learning and deep learning for classification and regression of hyperspectral data
• Embedded AI techniques for real-time hyperspectral image processing and target identification
• AI-based hyperspectral applications for food safety and quality assessment
• AI-based hyperspectral applications for field and controlled-environment agriculture
Hyperspectral imaging (HSI) is a powerful technique capable of obtaining both spatial and spectral information from a target by combining conventional machine vision and point spectroscopy methods. During the past two decades, the HSI technologies have found numerous successful applications for food and agriculture. Owing to growth in data, advances in machine learning algorithms and software, and increased computing power, artificial intelligence (AI) is currently driving a new wave of innovation that is poised to revolutionize many industries. Sensing technologies in various disciplines are becoming one of primary frontiers for the AI-driven developments and transformations. By leveraging the combination of the HSI and learning-based AI techniques, it is promising to accomplish novel sensing applications with measurement capabilities and accuracies that were not possible in the past.
This Research Topic intends to collect papers that highlight recent advances in research and development of AI-based hyperspectral imaging methodologies and their applications in the food and agricultural areas. Recently, miniature imaging spectrographs/spectrometers, all-in-one small hyperspectral cameras, and snapshot hyperspectral imagers are commercially available, which make it possible to develop compact and portable imaging systems and devices that can be used in confined space (e.g., overhead imaging in plant growth chambers) and field applications (e.g., crop health/disease sensing using drones or ground platforms). Learning-based data analysis methods, along with high-performance computing, open a new avenue to process hyperspectral images for classification and regression purposes. Integration of AI capabilities through machine learning into hyperspectral imaging systems is becoming a new axis to develop novel smart sensing devices and instruments. Articles presenting the development of compact and portable spectral imaging systems, multimodal hyperspectral imaging and data fusion methods, advanced machine learning and deep learning algorithms for hyperspectral data analysis, embedded AI techniques for real-time inspections, as well as AI-based hyperspectral applications for food, precision agriculture, and high-throughput phenotyping are of particular interest.
This Research Topic welcomes Original Research, Review, Mini Review, and Perspective manuscripts on themes including but not limited to:
• Compact and portable hyperspectral and multispectral systems and devices for smart sensing
• Single and multimodal hyperspectral imaging and data fusion (e.g., reflectance, fluorescence, Raman, and terahertz)
• Machine learning and deep learning for classification and regression of hyperspectral data
• Embedded AI techniques for real-time hyperspectral image processing and target identification
• AI-based hyperspectral applications for food safety and quality assessment
• AI-based hyperspectral applications for field and controlled-environment agriculture