Optimized the Spectral Reflectance Database of Herbal Plant Sample through ASD FieldSpc4 Spectroscopy

Abstract

Unique spectral signatures are important for reflecting the inherent characteristics of materials. But different factors can alter the spectral signature, thus affecting measurement quality. Environmental and experimental conditions, atmospheric-properties, timing, orientation, height of measurement, FOV, calibration of spectral data and spectral averaging are some of the main factors that affect measurements. This paper presents a database collection method of performing spectral reflectance. The procedure involves calibration protocols, data acquisition routines, preprocessing methods, and storage in a structured database for retrieval and analysis efficiency. The spectral data are classified into plant species, environmental conditions, and measurement parameters for increased usability in machine learning and remote sensing studies. The derived database is a rich source of information for researchers in vegetation classification, stress identification, and hyperspectral modeling. Measuring the spectral signature using ASD fieldspec4 spectroradiometer hyperspectral non-imaging device. The research focus on the importance of optimized spectral data collection and management to aid new researcher’s decision making in environmental monitoring.

Introduction

Hyperspectral remote sensing plays a crucial role in various fields including agriculture, medicine, and industry. The ASD FieldSpec4 spectroradiometer is a widely used instrument for capturing spectral reflectance data across the 350–2500 nm range, covering the Visible (VIS), Near-Infrared (NIR), and Shortwave Infrared (SWIR) regions. It provides high-resolution data (2151 bands) suitable for detailed analysis of materials, especially plant leaves, which are key identifiers due to their unique spectral signatures.

Spectral Data Collection Process

• The study was conducted at Dr. Babasaheb Ambedkar Marathwada University, Maharashtra.

• Six plant species were selected, with 9 leaves per plant (3 each of large, medium, and small size).

• Each leaf underwent 10 front and 10 back scans, resulting in 1080 total scans.

• Data collection included metadata: botanical name, local name, date, time, location, and environmental conditions.

• Leaves were stored in airtight bags before scanning.

Instrument Details

• The FieldSpec4 spectroradiometer includes:

  • 1.5 m fiber optic cable

  • Multiple fields of view (1°, 8°, 25°)

  • Laptop connectivity via Wi-Fi or Ethernet

  • Fast scanning (100 ms), high spectral resolution, and low stray light

Measurement Procedures

1. Warm-Up Period: Required for thermal stability (30–90 minutes).

2. Initial Setup: RS3 software is used for controlling and viewing spectral data.

3. Configuration: Recommended sample averaging = 10 for balance between noise and response time.

4. Device Optimization: Ensures proper light source settings; re-optimize if drift or saturation occurs.

5. White Reference Calibration: A spectralon panel provides a 100% reflectance baseline. It should be read every 10–15 minutes during scanning.

Applications

This hyperspectral approach is useful for:

• Plant species identification

• Environmental monitoring

• Soil and vegetation analysis

• Industrial and defense applications

Conclusion

This paper outlines the fundamental criteria for establishing a standard hyperspectral database using the ASD FieldSpec-4 Spectroradiometer, covering a range from 350nm to 2500nm. It encompasses a broad spectrum of research applications, such as crop classification, soil classification, identification of medicinal plants, and more, relying on spectral band variations. The ASD instrument facilitates laboratory and field database collection, offering options for 1o, 8 o, and 25oFOV based on the diameter of the target. The standard approach to database collection is grounded in environmental parameters.

References

[1] S. Kavitha, T. Satish Kumar, E. Naresh, Vijay H. Kalmani, Kalyan Devappa Bamane, Piyush Kumar Pareek, “Medicinal Plant Identifcation in Real Time Using Deep Learning Model”, SN Computer Science, 7 December 2023. [2] Adams Begue, Venitha Kowlessur, Fawzi Mahomoodally, Upasana Singh, Sameerchand Pudaruth, “Automatic Recognition of Medicinal Plants using Machine Learning Techniques”, International Journal of Advanced Computer Science and Applications, Vol. 8, No. 4, 2017. [3] Yousef Sharrab, Dimah Al-Fraihat, MontherTarawneh, Ahmad Sharieh, “Medicinal Plants Recognition Using Deep Learning”, International Conference on Multimedia Computing, Networking and Applications (MCNA), IEEE, 24 July 2023. [4] Gaju Chavan, Sonali Kulkarni “Identification of Plant Species using Remote Sensing Techniques: A Review”, High Technology Letters, Volume 29, Issue 9, 2023. [5] Pooja Vinod Janse, Jaypalsing N. Kayte, Rashmi V. Agrawal, Ratnadeep R. Deshmukh, “Standard Spectral Reflectance Measurements for ASD FieldSpec Spectroradiometer”, IEEE, Solan, India PDGC-2018 [6] Ghada Khdery, Mona Yones, “Innovative spectral library for identification common wild plants using hyperspectral technology in Northwestern Coast”, Egypt, Elsevier, 2020. [7] KalananthniPushpanathan, Marsyita Hanafi, SyamsiahMashohor, “Wan FazilahFazlil Ilahi, Machine learning in medicinal plants recognition: a review”, IEEE, 2020 [8] Manojkumar P., Surya C. M., Varun P. Gopi, “Identification of Ayurvedic Medicinal Plants by Image Processing of Leaf Samples”, IEEE, 2017. [9] Marcos Jiménez and Ricardo Díaz-Delgado, “Towards a Standard Plant Species Spectral Library Protocol for Vegetation Mapping: A Case Study in the Shrubland of Doñana National Park”, ISPRS Int. J. Geo-Inf. 2015, 4, 2472- 2495; doi:10.3390/ijgi4042472, 2015. [10] Kishore M, Dr. S. B. Kulkarni, “Hyperspectral Imaging Technique for Plant Leaf Identification”, International Conference on Emerging Research in Electronics, Computer Science and Technology, IEEE, 2015. [11] Pradip Salve, Pravin Yannawar, Milind Sardesai, “Multimodal plant recognition through hybrid feature fusion technique using imaging and non-imaging hyper-spectral data”, Elsevier B.V. on behalf of King Saud University, 2022. [12] Prabhat Kumar Thella, Dr.V. Ulagamuthalvi, “A Brief Review on Methods for Classification of Medical Plant Images”, IEEE, 2021. [13] Rupali R. Surase, AmarsinhVarpe, Mahesh Solankar, Hanumant Gite, Karbhari Kale, “Development of Nonimaging Spectral Library via Field Spec4 Spectroradiometer”, IJREAM, 2018.

Copyright

Copyright © 2025 Gaju S. Chavan, Simran N. Maniyar, Sonali B. Kulkarni. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.