Multi-scale advanced approaches to high-throughput phenotyping in crop improvement

Introduction

The collection of phenotypic data for a large number of phenotypic characters in a large extensive area is labor-intensive and challenging. Technological advancements have given rise to the concept of high-throughput phenotyping methods. High-throughput phenotyping (HTP) rapidly measures multiple plant traits without harm. It tracks growth, yield and stress responses using non-destructive methods (Abebe et al., 2023; Pabuayon et al., 2019) [2,6] . Plant phenotyping is a more recent and comprehensive definition for plant phenotyping, which assesses complex plant traits such as growth, development, tolerance, resistance, architecture, physiology, ecology, yield (Cadle-Davidson et al., 2019; Chaitra et al., 2017) [15,18] and the basic measurement of individual quantitative parameters that form the basis for complex trait assessment. These technologies enable researchers to collect vast amounts of phenotypic data quickly and efficiently, allowing for more comprehensive analyses of plant traits (Sheikh et al., 2023; He et al., 2024) [87,45] . These advanced methods can capture subtle differences in plant characteristics that may be overlooked by traditional phenotyping approaches. By integrating HTP with other cutting-edge technologies, scientists can gain deeper insights into plant biology and accelerate crop improvement efforts (Casto et al., 2021; Sheikh et al., 2023) [17,87] . High-throughput phenotyping has been implemented primarily to address the following challenges: outdated phenotyping tools, accelerated genomics technologies -a large gap between genotype and phenotype, deriving new traits that were not previously considered for phenotyping on the whole population, dynamic phenotyping, automation and robotics -increase accuracy and non-destructive phenotyping (Chen et al., 2015; Choudhury, 2020; Namin et al., 2017) [21,23,65] . The benefits arising from the utilization of HTP technologies facilitate the identification of novel traits that may have been overlooked using traditional phenotyping methods, potentially leading to new breeding targets and improved crop varieties (Sheikh et al., 2023) [87] . These technologies offer a solution to bridge the gap between rapidly advancing genomic technologies and traditional phenotyping methods. Providing more accurate and detailed phenotypic data enables researchers to correlate genotypic information with observable plant traits (Kiranmai, 2023; Nuzzo et al., 2009) [11,67] . This enhanced understanding of genotype-phenotype relationships can lead to more targeted and efficient crop breeding strategies, ultimately accelerating the development of improved varieties. These advanced techniques also facilitate dynamic phenotyping, enabling researchers to monitor plant traits over time and in response to various environmental conditions (Dalal et al., 2020; Attia et al., 2020) [26,10] . This temporal and contextual information can provide valuable insights into plant development, stress responses, and adaptive mechanisms, thereby enhancing our understanding of complex plant-environment interactions. Advanced phenotyping technologies not only improve accuracy through automation and robotics but also enable non-destructive measurements, allowing for repeated observations on the same plants throughout their growth cycle (Van De Zedde, 2022) [97] . This noninvasive approach provides a more comprehensive understanding of plant development and responses to environmental factors, thereby facilitating more informed breeding decisions. The integration of these technologies with machine-learning algorithms can further enhance data analysis, elucidating subtle patterns and relationships that may not be apparent through traditional methods (Gupta et al., 2016; Pathak et al., 2023; Pedro, 2023) [41, 73, 74] .

High throughput phenotyping

High-throughput phenotyping (HTP) technologies are revolutionizing crop improvement by enabling precise measurements of various traits across thousands of field-grown plants in diverse environments (Shakoor et al., 2017; Jangra et al., 2021; Sheikh et al., 2023) [84,49,87] . This approach is critical for selecting superior lines based on yield, disease resistance, and stress tolerance, thereby accelerating breeding programs aimed at enhancing crop performance (Paliwal et al., 2021; Sreenivasulu et al., 2006; Subiramani et al., 2020; Thakkar et al., 2021) [70,90,92] .

Understanding phenomics

The term "phenomics," introduced by Gerlai in 2002, refers to the comprehensive study of phenotypes at multiple biological levels, from molecular to organismal scales (Biase, 2023; Chen et al., 2013) [14,22] . High-throughput methods facilitate the analysis of plant growth, architecture, performance and composition, employing advanced imaging technologies such as hyperspectral imaging and 3D scanning. These non-invasive techniques allow researchers to capture detailed morphological and physiological data, significantly enhancing the understanding of plant responses to environmental stimuli (Haase, 2012) [42] .

Applications of high-throughput phenotyping

High throughput phenotyping technologies support the simultaneous analysis of multiple traits across various genetic lines and environmental conditions (He et al., 2024; Morota, 2024) [45,64] . This capability not only expedites the identification of desirable traits but also enriches the understanding of gene functions and interactions within the context of plant responses to stressors. By integrating phenomic data with other omics approaches, researchers can explore the intricate relationships between genotype, phenotype and environment, which is essential for developing targeted crop improvement strategies (Yang et al., 2021; Zaghlool & Attallah, 2022) [103,105] . Moreover, high-throughput phenotyping is vital for evaluating mutant populations, mapping populations, breeding populations, and germplasm collections. The development of phenotyping devices, including local cameras and infrared thermography, has enabled the assessment of stomatal opening and osmotic stress, providing quantitative data that can detect germplasm resistant to abiotic stresses in various plant species, including cereals and Arabidopsis. These phenotyping technologies are indispensable for modern plant breeding, offering a pathway to enhance crop resilience and productivity (Zhou et al., 2021) [109] . By facilitating comprehensive analyses of plant traits in varied growth conditions, these technologies not only streamline the breeding process but also impact the development of crops better suited for future agricultural challenges.

Basics to vision-based plant phenotyping

To know the basics of plant phenotyping we need to consider two main aspects, one is the image type acquired and how to process it and extract information from it. So, we are covering the current and emerging methods of image acquisition and processing allowing image-based phenomics. The high throughput phenotyping is based on images captured using different devices based on different light wavelength ranges and processing them and extracting the information (Danilevicz et al., 2021; Tiwari et al., 2021) [27,95] . So, before understanding different types of phenotyping methods available it is necessary to understand the step-by-step process of extracting this phenotypic information from images and different softwares and cameras used has to be known to the scientist. The process of extracting phenotypic information from images involves multiple steps, including image acquisition, preprocessing, segmentation, feature extraction, and data analysis (Guo et [39,51,82] . Various specialized software tools and algorithms have been developed to facilitate these steps, enabling researchers to efficiently process large volumes of plant images (Knecht et al., 2016) [82] . Understanding the capabilities and limitations of different imaging devices, such as multispectral cameras, thermal imagers, and hyperspectral sensors, is crucial for selecting the most appropriate technology for specific phenotyping applications.

Source: Yao et al., 2021 [104] [60,8] . Recent technological advancements in cameras indicate that manufacturers such as IMEC (India-Middle East-Europe Economic Corridor), a world leader in nanoelectronics, are working to integrate TDI (time delay integration) technology with image sensor characteristics within a single device. TDI technology is anticipated to be applied to high-throughput phenotyping processes in the near future (Bai & Ge, 2021) [12] . Image Acquisition Systems Classification: Image acquisition systems can be categorized into seven groups that are applicable for phenotyping: Image Analysis: Image analysis covers all stages from preprocessing to machine learning in plant image analysis (Zhang & Souri, 2022) [107] . The extraction of information from images is undertaken through the process of segmentation. The objective of a segmentation procedure is to isolate the components of an image that are of interest, i.e., the object or region of interest, from the remainder of the image, i.e., the background or irrelevant components. Consequently, this results in a partitioned image with significant regions.

B. Image pre-processing:

Image preprocessing is a crucial component of image analysis. The objective of image preprocessing is to enhance contrast and eliminate noise to accentuate the objects of interest in a given image (Ariateja et al., 2018; Pandey et al., 2023) [9,71] . Preprocessing techniques can range from basic operations such as image cropping and contrast enhancement to more complex procedures such as dimensionality reduction via Principal Component Analysis or Clustering (Gang & Bajwa, 2021; Zhang et al., 2018) [35,106] . In a comparative study analysing leaf diseases, histogram equalization was determined to be the most effective method for preprocessing colour images converted to grayscale.

C. Image segmentation:

Image segmentation is the core of image processing for artificial vision-based plant phenotyping (Choudhury, 2020; Patel, 2024) [23,72] . Segmentation facilitates the isolation and identification of objects of interest from an image and it aims to discriminate background or irrelevant objects (Patel, 2024) [72] . The objects of interest are defined by the internal similarity of pixels in parameters such as texture, colour, statistics, etc. One of the fundamental algorithms utilized is threshold segmentation, based on creating groups of pixels on a grayscale according to the level of intensity, thus separating the background from targets. This approach has been implemented with Android OS (Ap Leaf) to identify plant leaves (Wang et al., 2013) [99] . i) Otsu Method: Minimizes within-class variance to automatically segment images, frequently employed for background subtraction in plant recognition. ii) Watershed Transformation: Segments images by treating them as topological surfaces, terminating at strong edges. iii) Grab cut: Separates foreground from background using graph theory, effective with simple backgrounds but encounters difficulties with complex ones. iv) Snakes: Active contour method that refines segmentation by fitting splines to edges, utilized in plant phenotyping with features such as colour intensity and texture. v) Active Contours: Employed for precise plant recognition by combining algorithms for image segmentation, particularly effective for flowers.

D. Features extraction:

Features extraction constitutes one of the fundamental components of object identification and classification based on computer vision. The features extracted from an image are organized in "feature vectors" (Kumaraguru & Chakravarthy, 2017; Qu, 2019) [54,77] . The primary features include edges, pixel intensity, geometries, textures, and image transformations. One proposed system utilizes a feature vector composed of a combination of RGB and CIE L*a*b* colour spaces to segment images captured during daylight hours (Jing et al., 2015) [50] .

For night-time image segmentation, a vector comprising statistical features over two decomposition levels of the wavelet transform using IR images was computed. Numerous algorithms exist to identify invariant feature detectors and descriptors. This type of image analysis ensures the detection of points of interest in a scale-and rotation-independent manner. [30,36] . CNN [1,81,96] . [44,99] . This technique represents the most basic imaging technology for plant sensing. Typically, raw image data is presented as spatial matrices of intensity values corresponding to photon fluxes in the red (~600 nm), green (~550 nm), and blue (~450 nm) spectral ranges of visible light. Cameras operating in the visible spectrum, such as standard digital cameras or RGB/CIR cameras, are commonly used due to their ability to provide quick measurements and cost-effective solutions for plant phenotyping applications (Nijland et al., 2013) [66] . This colour information give estimation of the degree of senescence. By examining the colors, we can find the senescence of older leaves during drought and then suggest an escape or avoidance. Genotypes with stay-green type can be detected that would be able to continue photosynthesis under water stress. It also measures the aspects of plant architecture such as image-based projected biomass, leaf disease severity assessments, seed morphology, root architecture, leaf area, colour, growth dynamics, seedling vigour, yield, and fruit number and distribution.

The factors essential for accelerating plant phenomics

3-Dimensional imaging

In this method firstly the digital photos of the plants on top and side views were taken and combined into 3D image (Guo et al., 2018) [38] . The different measurements taken using 3-D were Shoot mass, Leaf number, Shape and angle, Leaf colour and Leaf health which are useful for constructing the final image in 3-dimensional representation.

Near infrared imaging (NIR)

NIR spectroscopy is a method that makes use of the nearinfrared region of the electromagnetic spectrum (from about 700 to 2500 nanometers) ( Engelhardt & Gillam-Krakauer, 2012) [5,19,32] . NIR Capable of examining irregular surfaces with the same ease as a carefully prepared sample. NIR is non-destructive and requires little or no sample preparation. It can also be used to analyse multiple constituents in a single scan (Chelladurai & Jayas, 2014; Cozzolino, 2009) [20,25] . NIR reflectance can be influenced by leaf thickness in addition to leaf water content.

Source: https://midopt.com/healthy-crop/

Benefits of NIR:

The NIR and short-wave infrared (SWIR) subranges contains several major and minor water absorption troughs that can be used as an index of leaf relative water content, more the presence of chlorophyll more will be reflectance in NIR range, it also facilitates estimation of water content and movement within leaves and soil and Carbohydrate content of leaves.

Far-infrared imaging (FIR)

Thermal imaging allows for the visualization of infrared radiation, indicating an object as the temperature across the object's surface. Far-infrared (FIR) spectroscopy utilizes the electromagnetic spectrum's far-infrared region, ranging from approximately 10µm to 1 mm. Thermal cameras typically operate within a sensitive spectral range of 3-14 µm (Ozaki, 2021; Ruan et al., 2001), with the most frequently employed wavelengths being 3-5 µm or 7-14 µm. These cameras are capable of detecting long-wave infrared (LWIR) radiation emitted by objects based on their temperature.

Benefits of FIR

FIR technology offers several advantages, including the ability to measure temperature differences between leaves and plants, identify cooler plants that absorb more water, and assess individual plants or entire plant systems. The temperature variations observed can be used to evaluate photosynthetic activity, salinity tolerance, and effective water use efficiency.

Fluorescence imaging

Plant metabolic status can be assessed by artificially stimulating plant photosystems and observing their responses. Fluorescence, which is the emission of light during radiation absorption at shorter wavelengths, occurs when chloroplasts are exposed to [24,100] . UV (ultraviolet) illumination (340-360 nm range) produces two types of fluorescence: red to far-red and blue to green. This forms the basis of multicolour fluorescence imaging, which allows for the simultaneous capture of fluorescence emission from four spectral bands (blue (440 nm, F440), green (520 nm, F520), red (690 nm, F690), and far-red (740 nm, 740)) using a single excitation wavelength. The blue and green fluorescence signals (with peaks near 440 and 520 nm) originate from cinnamic acids, primarily ferulic acid, found mostly in cell walls (Fincher, 1976; Lichtenthaler & Schweiger, 1998; Morales et al., 1996) [33,55,63] . The red and far-red fluorescence emissions (with peaks near 690 and 740 nm) come from chlorophyll α molecules in the antenna and reaction centre of photosynthetic photosystem II, located in chloroplasts within mesophyll cells (Van Dorssen et al., 1987).

Benefits of FI

It measures photosynthesis rate, biotic and abiotic stress responses and chlorophyll content.

Hyperspectral imaging

Hyperspectral imaging combines spectral (λ) and spatial (x, y) data into a three-dimensional structure known as a 'hyperspectral data cube' or 'hypercube' (Sharma et al., 2021; Xu et al., 2016) [21, 102] . This 3D hypercube consists of two dimensions of spatial information and one spectral dimension, which contains data for numerous spectral bands.

Sonar

An ultrasonic sensor is an electronic device that determines the distance to a target object by sending out ultrasonic sound waves and transforming the reflected sound into an electrical signal (Canali et al., 1982; Kulkarni et al., 2019; Punia, 2014; Srijith, 2021; Tanaka, 1987) [16,53,76,91] . A CTFM ultrasonic sensor generates a signal containing information about the geometric structure of plants (Darwito et al., 2019; Dorr, 1991; Gupta & Agarwal, 2018) [28,31,40] . By correlating echoes from various angles, plants can be identified with enough precision for navigation purposes ( [7,47,83] , has emerged as an innovative active sensing technology for three-dimensional measurement of plant morphology and canopy structures. This technology accurately determines the distance between the sensor and a target by measuring either the time elapsed between laser pulse emission and return (the 'time of flight' method) or through trigonometric calculations (the 'optical probe' or 'light section' methods) (Fu et al., 2011; Hardesty, 1991) [34,43] . Airborne lidar systems typically achieve accuracies of ∼0.1-1 m, while ground-based systems offer precision ranging from ∼0.05-10 cm, making lidar a superior alternative to traditional passive 3D measurement techniques. Various categories of lidar systems exist, classified according to their specific characteristics.

Computed tomography (CT)

X-ray digital radiography serves as the foundation for this imaging technique. It is commonly employed to evaluate tissue density, count tiller numbers, and assess grain quality, among other applications. Nevertheless, the processes of CT scanning, reconstruction, and extracting roots from X-ray CT volumes are time-consuming (Michael, 2001) [61] . For genetic studies, such as quantitative trait locus mapping, which necessitate large population sizes, a method that can visualize RSA rapidly and efficiently is essential.

Magnetic resonance imaging (MRI)

This imaging method aims to display metabolites, offer structural insights, and track internal physiological activities occurring within living organisms (Wirestam, 2021) [101] . This innovative approach allows researchers to visualize and analyse the distribution of various biochemical compounds within the body. By providing detailed metabolic mapping, it enables a deeper understanding of tissue composition and physiological processes. Moreover, this technique offers the potential to monitor real-time changes in metabolic activity, which could be invaluable for diagnosing and tracking disease progression.

Ground penentrating RADAR

Ground Penetrating Radar (GPR) is a non-invasive geophysical method that uses electromagnetic waves to detect subsurface objects, structures, and changes in material properties (Ismail et al., 2016; Slob et al., 2010) [48,89] . It is commonly used in archaeology, geology, and civil engineering to map underground features without excavation. GPR systems emit high-frequency radio waves into the ground, which reflect back upon hitting different materials, creating detailed images of subsurface layers. Its applications range from locating buried utilities to assessing soil conditions and detecting hidden artifacts.

Future prospects:

The future of high-throughput phenotyping (HTP) in plant science holds great promise, propelled by rapid advancements in sensor technologies, artificial intelligence, and data analytics, which are transforming the way we assess plant traits on a large scale (Ma et al., 2022; Sharma & Shivandu, 2024) [56,86] . These innovations, including mobile phenotyping platforms, UAVs, and satellite imaging, are enhancing the precision and efficiency of phenotypic data collection across diverse environments. As HTP methods evolve, the integration of multi-omics data, including genomic and metabolomic information, is expected to provide deeper insights into plant responses to environmental challenges, aiding the development of more resilient crop varieties. However, to fully harness the potential of HTP, the field must address key challenges, such as standardizing phenotyping protocols, managing the vast amounts of data generated, and overcoming the high costs of infrastructure, particularly for researchers in resource-limited settings. Addressing these challenges will be crucial in leveraging HTP to advance crop breeding, especially in the face of climate change.

Conclusion

In conclusion, the future of high-throughput phenotyping is bright, with technological advancements and innovative approaches paving the way for improved agricultural practices.

Addressing existing challenges and focusing on integration and automation will be key to unlocking the full potential of these methods in plant science.

References

1. Abade A, Ferreira P, Vidal F. Plant diseases recognition on images using convolutional neural networks: A systematic review. Cornell University; c2020. Available from: https://doi.org/10.48550/arxiv.2009.04365
2. Abebe AM, Baek J, Kim Y, Kim J, Kim SL. Image-based high-throughput phenotyping in horticultural crops. Plants. 2023;12(10):2061. Available from: https://doi.org/10.3390/plants12102061
3. Adebola Olayinka A, Adewale Oluwadamilare A, Femi Emmanuel A. Distance measurement and energy conservation using Arduino Nano and ultrasonic sensor. American Journal of Electrical and Computer Engineering. 2021;5(2):40. Available from: https://doi.org/10.11648/j.ajece.20210502.11
4. Agarwal A, Samaiya D, Gupta KK. A comparative study of SIFT and SURF algorithms under different object and background conditions. Institute of Electrical and Electronics Engineers; c2017. Available from: https://doi.org/10.1109/icit.2017.48
5. Ali J, Ansari S, Pramod K. Near-infrared spectroscopy for nondestructive evaluation of tablets. Systematic Reviews in Pharmacy. 2010;1(1):17. Available from: https://doi.org/10.4103/0975-8453.59508
6. Araus JL, Kefauver SC, Zaman-Allah M, Olsen MS, Cairns JE. Translating high-throughput phenotyping into genetic gain. Trends in Plant Science. 2018;23(5):451-166. Available from: https://doi.org/10.1016/j.tplants.2018.02.001
7. Archibald C, Petriu E. Robot skills development using a laser range finder. Institute of Electrical and Electronics Engineers; c1993. Available from: https://doi.org/10.1109/imtc.1993.382601
8. Ardeshirpour Y, Deen MJ, Shirani S. Evaluation of complementary metal-oxide semiconductor based photodetectors for low-level light detection. Journal of Vacuum Science and Technology A. 2006;24(3):860-865. Available from: https://doi.org/10.1116/1.2190652
9. Ariateja D, Soesanti I, Ardiyanto I. A review of contrast enhancement techniques in digital image processing. Institute of Electrical and Electronics Engineers; 2018. Available from: https://doi.org/10.1109/icstc.2018.8528579
10. Attia Z, Grunwald Y, Shenhar I, Bourstein R, Averbuch N, Wallach R, et al. A telemetric, gravimetric platform for real-time physiological phenotyping of plant-environment interactions. Journal of Visualized Experiments. 2020;(162). Available from: https://doi.org/10.3791/61280
11. B Kiranmai EA. Comparative analysis association and prediction of various phenotypic traits of Oryza sativa. International Journal of Recent Innovations in Computing and Communication. 2023;11(10):1471-1480. Available from: https://doi.org/10.17762/ijritcc.v11i10.8697
12. Bai G, Ge Y. Cable suspended large-scale field phenotyping facility for high-throughput phenotyping research. Springer; 2021. p. 39-53. Available from: https://doi.org/10.1007/9783-030-73734-4_3
13. Bakhshi Y, Kaur S, Verma P. An efficient approach in face recognition for invariant faces using SIFT, SURF and PCA. International Journal of Signal Processing, Image Processing and Pattern Recognition. 2016;9(5):99-108. Available from: https://doi.org/10.14257/ijsip.2016.9.5.10
14. Biase F. Molecular phenomics-a multi-omics approach to understanding complex traits in cattle. Journal of Animal Science. 2023;101(Suppl 3):53. Available from: https://doi.org/10.1093/jas/skad281.066
15. Cadle-Davidson L, Sapkota S, Gutierrez B, Martinez D, Londo J. From phenotyping to phenomics: Present and future approaches in grape trait analysis to inform grape gene function. Springer; 2019. p. 199-222. Available from: https://doi.org/10.1007/978-3-030-18601-2_10
16. Canali C, De Cicco G, Prudenziati M, Taroni A, Morten B. A temperature compensated ultrasonic sensor operating in air for distance and proximity measurements. IEEE Transactions on Industrial Electronics. 1982;IE-29(4):33641. Available from: https://doi.org/10.1109/tie.1982.356688
17. Casto AL, Tovar JC, Schuhl H, Wang Q, Fahlgren N, Bart RS, et al. Picturing the future of food. Plant Phenomics Journal. 2021;4(1). Available from: https://doi.org/10.1002/ppj2.20014
18. Chaitra N, Veena S, Raj T, Parinitha J. A survey on plant phenotype. Indian Journal of Science and Technology. 2017;10(4). Available from: https://doi.org/10.17485/ijst/2017/v10i4/110713
19. Chapman J, Elbourne A, Truong VK, Cozzolino D. Shining light into meat-a review on the recent advances in in vivo and carcass applications of near infrared spectroscopy. International Journal of Food Science and Technology. 2019;55(3):935-941. Available from: https://doi.org/10.1111/ijfs.14367
20. Chen CY, Wang ML, Dang PM, Butts CL. Advances in phenotyping of functional traits. Springer India; 2015. p. 163-80. Available from: https://doi.org/10.1007/978-81322-2226-2_11
21. Colarusso P, Spring KR. Imaging at low light levels with cooled and intensified charge-coupled device cameras. Methods in Enzymology. 2003;360:383-394. Available from: https://doi.org/10.1016/s0076-6879(03)60120-5
22. Cozzolino D. Near infrared spectroscopy in natural products analysis. Planta Medica. 2009;75(7):746-756. Available from: https://doi.org/10.1055/s-0028-1112220
23. Dalal A, Bourstein R, Mayo A, Shenhar I, Wallach R, Grunwald Y, et al. A high-throughput gravimetric phenotyping platform for real-time physiological screening of plant-environment dynamic responses. Cold Spring Harbor Laboratory; c2020. Available from: https://doi.org/10.1101/2020.01.30.927517
24. Danilevicz MF, Edwards D, Nestor BJ, Bayer PE, Bennamoun M. Resources for image-based high-throughput phenotyping in crops and data sharing challenges. Plant Physiology. 2021;187(2):699-715. Available from: https://doi.org/10.1093/plphys/kiab301
25. Darwito PA, Sa'Diyah H, Aditya W, Cikadiarta A, Raditya M. Comparative study of burst and beams types ultrasonic sensor for distance measurements. Institute of Electrical and Electronics Engineers; c2019. Available from: https://doi.org/10.1109/isitia.2019.8937133
26. Dhaka VS, Ijaz MF, Sinwar D, Woźniak M, Meena SV, Kavita K, et al. A survey of deep convolutional neural networks applied for prediction of plant leaf diseases. Sensors (Basel). 2021;21(14):4749. Available from: https://doi.org/10.3390/s21144749
27. Dorr JA. Ultrasonic measuring system. The Journal of the Acoustical Society of America. 1991;90(1):622. https://doi.org/10.1121/1.401218
28. Engelhardt B, Gillam-Krakauer M. Use of Near-Infrared Spectroscopy in the Management of Patients in Neonatal Intensive Care Units -An Example of Implementation of a New Technology. Institute for New Technologies. 2012. https://doi.org/10.5772/37994
29. Fincher GB. Ferulic acid in barley cell walls: a fluorescence study. Journal of the Institute of Brewing. 1976;82(6):347349. https://doi.org/10.1002/j.2050-0416.1975.tb06961.x
30. Fu R, Cheng Y, Su C, Li M, Chang B, Guo Y. The research of the laser facula of laser ranger finder in the far distance. Society of Photo Optical Instrumentation Engineers. 2011;7912. https://doi.org/10.1117/12.873565
31. Gang A, Bajwa W. FAST-PCA: A Fast and Exact Algorithm for Distributed Principal Component Analysis. Cornell University. 2021. https://doi.org/10.48550/arxiv.2108.12373
32. Ghongade MA, Nagpur D. Plant Disease Detection Using Deep Learning-Based Convolutional Neural Networks With Transfer Learning Algorithms. International Journal of Scientific Research in Engineering And Management. 2024;08(05):1-5. https://doi.org/10.55041/ijsrem34775
33. Gill T, Gill SK, Saini DK, Chopra Y, de Koff JP, Sandhu KS. A comprehensive review of high throughput phenotyping and machine learning for plant stress phenotyping. Frontiers in Plant Science. 2022;13:959050. https://doi.org/10.3389/fpls.2022.959050
34. Guo J, Zhang X, Cheng Z, Jaeger M, Xu S, Yan D-M. Realistic Procedural Plant Modeling from Multiple View Images. IEEE Transactions on Visualization and Computer Graphics. 2018;26(2):1372-1384. https://doi.org/10.1109/tvcg.2018.2869784
35. Guo X, Gao G, An J, Cheng J. Research Algorithm on the Wheat Phenotypic Feature Extraction based on Image Processing. Journal of Physics: Conference Series. 2021;2010(1):012002. https://doi.org/10.1088/1742-6596/2010/1/012002
36. Gupta N, Agarwal AK. Object Identification using Super Sonic Sensor: Arduino Object Radar. Institute of Electrical Electronics Engineers. 2018. https://doi.org/10.1109/sysmart.2018.8746951
37. Gupta P, Jindal R, Sharma A. Scalable machine-learning algorithms for big data analytics: a comprehensive review. WIREs Data Mining and Knowledge Discovery. 2016;6(6):194-214. https://doi.org/10.1002/widm.1194
38. Haase DL. Using Nomograms for Evaluating Plant Morphological and Physiological Data. Western Journal of Applied Forestry. 2012;27(1):42-45. https://doi.org/10.1093/wjaf/27.1.42
39. Hardesty RM. Recent Advances in Coherent Laser Radar. Optica. 1991. https://doi.org/10.1364/orsa.1991.omc2
40. Harvey SGP, Bähr JL. A temperature-controlled complementary metal oxide semiconductor camera. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture. 2004;218(9):12171222. https://doi.org/10.1243/0954405041897112
41. He S, Yang W, Liu W, Chen M, Xu X, Gong T, et al. Crop HTP Technologies: Applications and Prospects. Agriculture. 2024;14(5):723. https://doi.org/10.3390/agriculture14050723
42. Hruška J, Morais R, Adão T, Marques P, Cunha A, Peres E, et al. Machine learning classification methods in hyperspectral data processing for agricultural applications. Association for Computing Machinery. 2018. https://doi.org/10.1145/3220228.3220242
43. Iman HB, Md Rashid NKA. Probabilistic model of laser range finder for three dimensional grid cell in close range environment. IIUM Engineering Journal. 2016;17(1):63-82. https://doi.org/10.31436/iiumej.v17i1.570
44. Ismail NA, Mohd Muztaza N, Saad R. Reflectivity of electromagnetic (EM) wave in shallow ground penetrating radar (GPR) survey. Jurnal Teknologi. 2016;78(7-3). https://doi.org/10.11113/jt.v78.9500
45. Jangra S, Yadav NR, Yadav RC, Chaudhary V. HighThroughput Phenotyping: A Platform to Accelerate Crop Improvement. Phenomics. 2021;1(2):31-53. https://doi.org/10.1007/s43657-020-00007-6
46. Jing J, Liu S, Li P, Zhang L. The fabric defect detection based on CIE Lab* color space using 2-D Gabor filter. The Journal of The Textile Institute. 2015;107(10):1305-1313. https://doi.org/10.1080/00405000.2015.1102458
47. Khobragade S, Mor DD, Chhabra A. A method of ear feature extraction for ear biometrics using MATLAB. Institute of Electrical Electronics Engineers. 2015. https://doi.org/10.1109/indicon.2015.7443344
48. Knecht AC, Swanson DR, Campbell MT, Walia H, Caprez A. Image Harvest: an open-source platform for highthroughput plant image processing and analysis. Journal of Experimental Botany. 2016;67(11):3587-3599. https://doi.org/10.1093/jxb/erw176
49. Kulkarni AU, Hegde S, Potdar AM, Baligar VP. RADAR based Object Detector using Ultrasonic Sensor. Institute of Electrical Electronics Engineers; c2019. https://doi.org/10.1109/icait47043.2019.8987259
50. Kumaraguru PV, Chakravarthy VJ. An Image Feature Extraction and Image Representation Strategies for the Analysis of Image Processing. Indian Journal of Forensic Medicine & Toxicology. 2017;11(2):642. https://doi.org/10.5958/0973-9130.2017.00202.x
51. Lichtenthaler HK, Schweiger J. Cell wall bound ferulic acid, the major substance of the blue-green fluorescence emission of plants. Journal of Plant Physiology. 1998;152(2-3):272-282. https://doi.org/10.1016/s0176-1617(98)80142-9
52. Ma Z, Liu Z, Xiao G, Ruan Y, Rayhana R, Feng K, Sangha JS. A Review on Sensing Technologies for HighThroughput Plant Phenotyping. IEEE Open Journal of Instrumentation and Measurement. 2022;1:1-21. https://doi.org/10.1109/ojim.2022.3178468
53. Magori V, Vossiek M, Ruser H, Eccardt PC. Directionsensitive ultrasonic distance sensor using multimode stimulation of a single transducer; c1995. https://doi.org/10.1109/ultsym.1995.495742
54. Mahlein AK, Steiner U, Dehne HW, Oerke EC. Highthroughput field-phenotyping tools for plant breeding and precision agriculture. Agronomy. 2019;9(5):258. https://doi.org/10.3390/agronomy9050258
55. Mayerich D, Keyser J, Choe Y, Kwon J, Panchal A. Fast cell detection in high-throughput imagery using GPUaccelerated machine learning. Institute of Electrical Electronics Engineers; c2011. https://doi.org/10.1109/isbi.2011.5872507
56. Mehta S, Patel A, Mehta J. CCD or CMOS Image sensor for photography. Institute of Electrical Electronics Engineers. 2015. https://doi.org/10.1109/iccsp.2015.7322890.
57. Mishra P, Asaari MSM, Herrero-Langreo A, Lohumi S, Diezma B, Scheunders P. Close range hyperspectral imaging of plants: A review. Biosystems Engineering. 2017;164:49-67. https://doi.org/10.1016/j.biosystemseng.2017.09.009.
58. Morales F, Cerovic ZG, Moya I. Time-resolved blue-green fluorescence of sugar beet (Beta vulgaris L.) leaves. Spectroscopic evidence for the presence of ferulic acid as the main fluorophore of the epidermis. Biochimica et Biophysica Acta (BBA) -Bioenergetics. 1996;1273(3):25162. https://doi.org/10.1016/0005-2728(95)00153-0.
59. Morota G. 123 Phenomics enabled quantitative genetic modeling of complex traits. Journal of Animal Science. 2024;102(Suppl 2):26. https://doi.org/10.1093/jas/skae102.032.
60. Nijland W, De Jong R, De Jong SM, Wulder MA, Bater CW, Coops NC. Monitoring plant condition and phenology using infrared sensitive consumer grade digital cameras. Agricultural and Forest Meteorology. 2013;184:98-106. https://doi.org/10.1016/j.agrformet.2013.09.007.
61. Nuzzo A, Bellazzi R, Riva A. Phenotypic and genotypic data integration and exploration through a web-service architecture. BMC Bioinformatics. 2009;10(S12). https://doi.org/10.1186/1471-2105-10-s12-s5.
62. Ozaki Y. Infrared Spectroscopy-Mid-infrared, Nearinfrared, and Far-infrared/Terahertz Spectroscopy. Analytical Sciences. 2021;37(9):1193-212. https://doi.org/10.2116/analsci.20r008.
63. Pabuayon ILB, Ritchie GL, Sun Y, Guo W. Highthroughput phenotyping in cotton: a review. Journal of Cotton Research. 2019;2(1). https://doi.org/10.1186/s42397-019-0035-0.
64. Paliwal R, Singh G, Mir RR, Gueye B. Chapter 7 - Genomic-assisted breeding for abiotic stress tolerance in horticultural crops. In: Stress Tolerance in Horticultural Crops. Elsevier; 2021. p. 91-118. https://doi.org/10.1016/b978-0-12-822849-4.00007-3.
65. Patel D. Computer Vision and Image Segmentation. International Journal for Research in Applied Science and Engineering Technology. 2024;12(2):915-925. https://doi.org/10.22214/ijraset.2024.58479.
66. Pathak S, Taware S, Pawar A, Kulkarni S, Akkalkot A. A Survey on Machine Learning Algorithms for RiskControlled Algorithmic Trading. International Journal of Scientific Research in Science and Technology. 2023;106989. https://doi.org/10.32628/ijsrst523103163.
67. Pedro F. A Review of Data Mining, Big Data Analytics and Machine Learning Approaches. Journal of Computing and Natural Science. 2023;169-81. https://doi.org/10.53759/181x/jcns202303016.
68. Peng S, Wang J, Guo X, Yang W, Zhao C. High-throughput phenotyping: Breaking through the bottleneck in future crop breeding. Plant Science. 2021;311:110993. https://doi.org/10.1016/j.plantsci.2021.110993.
69. Punia S. Ultra sonic range finding for distance measuring in coal mining. International Journal of Research in Engineering and Technology. 2014;03(11):313-5. https://doi.org/10.15623/ijret.2014.0311051.
70. Qu Y-H. Application of the Computer Vision Technology in the Image Feature Extraction. In: Springer Nature; 2019. p. 351-6. https://doi.org/10.1007/978-3-030-15235-2_53.
71. Rehman TU, Mahmud MS, Chang YK, Jin J, Shin J. Current and future applications of statistical machine learning algorithms for agricultural machine vision systems. Computers and Electronics in Agriculture. 2018;156:585605. https://doi.org/10.1016/j.compag.2018.12.006.
72. Routray S, Ray AK, Mishra C. Analysis of various image feature extraction methods against noisy image: SIFT, SURF and HOG. Institute of Electrical Electronics Engineers. 2017. https://doi.org/10.1109/icecct.2017.8117846.
73. Ruan HD, Frost RL, Kloprogge JT, Duong L. Far-infrared spectroscopy of alumina phases. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 2001;58(2):265-72. https://doi.org/10.1016/s1386-1425(01)00532-7.
74. de Almeida AS, Abade A, Vidal F. Plant Diseases Recognition from Digital Images using Multichannel Convolutional Neural Networks. Scitepress Science Technology. 2019;2. https://doi.org/10.5220/0007383904500458.
75. Samanta S, Tiwari VK, Gour B, Utkarsha S. Fast Character Recognition Using Kohonen Neural Network. 2018. https://doi.org/10.1109/icacat.2018.8933627.
76. Schwarte RM, Singh UN. Breakthrough in multichannel laser-radar technology providing thousands of highsensitive lidar receivers on a chip. Society of Photo Optical Instrumentation Engineers. 2004;5575. https://doi.org/10.1117/12.573727.
77. Shakoor N, Lee S, Mockler TC. High throughput phenotyping to accelerate crop breeding and monitoring of diseases in the field. Current Opinion in Plant Biology. 2017;38:184-92. https://doi.org/10.1016/j.pbi.2017.05.006.
78. Sharma G, Singh K, Parmar A, Naveed SK. Chromatic Dispersion Based Wide-Band, Fiber-Coupled, Tunable Light Source for Hyperspectral Imaging. IEEE Access. 2021;9:50538-45. https://doi.org/10.1109/access.2021.3068617.
79. Sharma K, Shivandu SK. Integrating artificial intelligence and Internet of Things (IoT) for enhanced crop monitoring and management in precision agriculture. Sensors International. 2024;5:100292. https://doi.org/10.1016/j.sintl.2024.100292.
80. Sheikh M, Iqra F, Ambreen H, Pravin KA, Ikra M, Chung YS. Integrating artificial intelligence and high-throughput phenotyping for crop improvement. Journal of Integrative Agriculture. 2023;23(6):1787-1802. https://doi.org/10.1016/j.jia.2023.10.019.
81. Singh A, Ganapathysubramanian B, Sarkar S, Singh AK. Machine Learning for High-Throughput Stress Phenotyping in Plants. Trends in Plant Science. 2015;21(2):110-124. https://doi.org/10.1016/j.tplants.2015.10.015.
82. Slob E, Olhoeft G, Sato M. Ground-Penetrating Radar. Society of Exploration Geophysicists; 2010. p. 141-60. https://doi.org/10.1190/1.9781560802273.ch7
83. Sreenivasulu N, Sopory SK, Kavi Kishor PB. Deciphering the regulatory mechanisms of abiotic stress tolerance in plants by genomic approaches. Gene. 2006;388(1-2):1-13. https://doi.org/10.1016/j.gene.2006.10.009
84. Srijith B. Arduino based Distance Measurement Sensor using Ultrasonic Sensor. Int J Res Appl Sci Eng Technol. 2021;9(6):1789-1799. https://doi.org/10.22214/ijraset.2021.35346
85. Tanaka H. Ultra-sonic distance sensor system and method with correction feature for sensor value. J Acoust Soc Am. 1987;82(1):404. https://doi.org/10.1121/1.395464
86. Tugrul B, Elfatimi E, Eryigit R. Convolutional Neural Networks in Detection of Plant Leaf Diseases: A Review. Agriculture. 2022;12(8):1192. https://doi.org/10.3390/agriculture12081192
87. Van De Zedde R. Field robots for plant phenotyping. Burleigh Dodds Science; 2022. p. 153-78. https://doi.org/10.19103/as.2022.0102.08
88. Wang B, Brown D, Gao Y, Salle JL. Mobile plant leaf identification using smart-phones. Institute of Electrical and Electronics Engineers; c2013. https://doi.org/10.1109/icip.2013.6738910
89. Webb DJ, Brown CM. Epi-fluorescence microscopy. Methods Mol Biol. 2012;931:29-59. https://doi.org/10.1007/978-1-62703-056-4_2
90. Xu M, Wu Z, Qiao Y, Zhang Q, Zhou L, Shi X. Nearinfrared chemical imaging for quantitative analysis of chlorpheniramine maleate and distribution homogeneity assessment in pharmaceutical formulations. J Innov Opt Health Sci. 2016;09(06):1650002. https://doi.org/10.1142/s1793545816500024
91. Yang Y, Li J, Zhang J, Wu Y, Abdelaal WB, Saand MA, et al. Applications of Multi-Omics Technologies for Crop Improvement. Front Plant Sci. 2021;12(Suppl 1). https://doi.org/10.3389/fpls.2021.563953
92. Yao L, Van De Zedde R, Kowalchuk G. Recent developments and potential of robotics in plant ecophenotyping. Emerg Top Life Sci. 2021;5(2):289-300. https://doi.org/10.1042/ETLS20200275
93. Zaghlool SB, Attallah O. A review of deep learning methods for multi-omics integration in precision medicine. Institute of Electrical and Electronics Engineers; 2022. https://doi.org/10.1109/bibm55620.2022.9995099
94. Zhang R, Qu S, Du T. A principal component analysis algorithm based on dimension reduction window. IEEE Access. 2018;6:63737-47. https://doi.org/10.1109/access.2018.2875270
95. Zhang X, Souri A. Application of artificial intelligence recognition technology in digital image processing. Wireless Commun Mob Comput. 2022;2022:1-10. https://doi.org/10.1155/2022/7442639
96. Zheng YJ. Self-organization grouping for feature extraction and image segmentation. 1994. https://doi.org/10.1109/sipnn.1994.344977
97. Zhou J, Zhou J, Nguyen HT, Ye H. High-throughput crop phenotyping systems for controlled environments. Springer; 2021. p. 183-208. https://doi.org/10.1007/978-3-030-73734-4_9.