Skip to main content
SpringerLink
Log in

Towards Discrimination of Plant Species by Machine Vision: Advanced Statistical Analysis of Chlorophyll Fluorescence Transients

  • Original Paper
  • Published:
Journal of Fluorescence Aims and scope Submit manuscript

Abstract

Automatic discrimination of plant species is required for precision farming and for advanced environmental protection. For this task, reflected sunlight has already been tested whereas fluorescence emission has been only scarcely considered. Here, we investigated the discriminative potential of chlorophyll fluorescence imaging in a case study using three closely related plant species of the family Lamiaceae. We compared discriminative potential of eight classifiers and four feature selection methods to identify the fluorescence parameters that can yield the highest contrast between the species. Three plant species: Ocimum basilicum, Origanum majorana and Origanum vulgare were grown separately as well as in pots where all three species were mixed. First, eight statistical classifiers were applied and tested in simulated species discrimination. The performance of the Quadratic Discriminant Classifier was found to be the most efficient. This classifier was further applied in combination with four different methods of feature selection. The Sequential Forward Floating Selection was found as the most efficient method for selecting the best performing subset of fluorescence images. The ability of the combinatorial statistical techniques for discriminating the species was also compared to the resolving power of conventional fluorescence parameters and found to be more efficient.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Notes

  1. Somewhat smaller sets were used with the slower algorithms: with the automatic neural networks classifier (2,250 transients) and with the support vector classifier (1,350 transients) was used for both training and testing set for each species.

  2. The Fisher’s Linear Discriminant Classifier is closely related to the Linear Discriminant Classifier.

  3. One should note that the feature selection techniques investigated here do not analyze all potential combinations of fluorescence images and in this sense they are called suboptimal. We choose this approach to maintain feasibility with present high performance personal computers. Because of that unexpected time consumption it was impossible to apply optimal techniques such as branch and bound [48], however it would significantly increase the classification performance of the reduced feature sets.

  4. These pots were different from those used for classifier training.

Abbreviations

CCD:

Charge Coupled Device

ChlF:

Chlorophyll Fluorescence

F0 :

minimal fluorescence level of dark adapted plants when primary quinone accepter (QA) of Photosystem II is oxidized

FM :

maximal fluorescence level of dark adapted leaves measured when QA and the plastoquinone pool are reduced

FV=FM-F0 :

variable fluorescence

FLDC:

Fisher’s Linear Discriminant Classifier

IFS:

Individual Feature Selection

k-NN:

k-Nearest Neighbor Classifier

LDC:

Linear Discriminant Classifier

NEURC:

Automatic neural networks Classifier

NMC:

Nearest Mean Classifier

NN:

Nearest Neighbor Classifier

NPQ:

Non-Photochemical Quenching

QDC:

Quadratic Discriminant Classifier

qP:

Photochemical quenching

Rfd:

Fluorescence decrease ratio

SBS:

Sequential Backward Selection

SFFS:

Sequential Forward Floating Selection

SFS:

Sequential Forward Selection

SVC:

Support Vector Classifier

References

  1. Feyaerts F, van Gool L (2001) Multi-spectral vision system for weed detection. Pattern Recognit Lett 22(6–7):667–674. doi:10.1016/S0167-8655(01)00006-X

    Article  Google Scholar 

  2. Kavdir I (2004) Discrimination of sunflower, weed and soil by artificial neural networks. Comput Electron Agric 44(2):153–160. doi:10.1016/j.compag.2004.03.006

    Article  Google Scholar 

  3. Gerhards R, Christensen S (2003) Real-time weed detection, decision making and patch spraying in maize, sugar beet, winter wheat and winter barley. Weed Res 43(6):385–392. doi:10.1046/j.1365-3180.2003.00349.x

    Article  Google Scholar 

  4. Timmermann C, Gerhards R, Kuhbauch W (2003) The economic impact of site-specific weed control. Precis Agric 4(3):249–260. doi:10.1023/A:1024988022674

    Article  Google Scholar 

  5. Wang N, Zhang N, Dowell FE, Sun Y, Peterson DE (2001) Design of an optical weed sensor using plant spectral characteristics. Trans ASAE 44(2):409–419

    Google Scholar 

  6. El-Faki MS, Zhang N, Peterson DE (2000) Weed detection using color machine vision. Trans ASAE 43(6):1969–1978

    Google Scholar 

  7. Cho SI, Lee DS, Jeong JY (2002) Weed-plant discrimination by machine vision and artificial neural network. Biosystems Eng 83(3):275–280. doi:10.1006/bioe.2002.0117

    Article  Google Scholar 

  8. Franz E, Gebhardt MR, Unklesbay KB (1995) Algorithms for extracting leaf boundary information from digital images of plant foliage. Trans ASAE 38(2):625–633

    Google Scholar 

  9. Gerhards R, Kuhbauch W (1993) Dynamic decision-model for weed control methods in cereals by means of digital image analysis. J Agron Crop Sci 171(5):329–335. doi:10.1111/j.1439-037X.1993.tb00148.x

    Article  Google Scholar 

  10. Tian LF, Slaughter DC (1998) Environmentally adaptive segmentation algorithm for outdoor image segmentation. Comput Electron Agric 21(3):153–168. doi:10.1016/S0168-1699(98) 00037-4

    Article  Google Scholar 

  11. Lee WS, Slaughter DC, Giles DK (1999) Robotic weed control system for tomatoes. Precis Agric 1:95–113

    Article  Google Scholar 

  12. Haggar RJ, Stent CJ, Isaac S (1983) A prototype hand-held patch sprayer for killing weeds, activated by spectral differences in crop weed canopies. J Agric Eng Res 28(4):349–358. doi:10.1016/0021-8634(83)90066-5

    Article  Google Scholar 

  13. Felton WL, McCloy KR (1992) Spot spraying. Agric Eng 73(6):9–12

    Google Scholar 

  14. Shiraishi M, Sumiya H (1996) Plant identification from leaves using quasi-sensor fusion. J Manuf Sci Eng Trans ASME 118(3):382–387. doi:10.1115/1.2831041

    Article  Google Scholar 

  15. Vrindts E, De Baerdemaeker J (2000) Using spectral information for weed detection in field circumstances. AgEng 2000 Conference, Warwick, UK

    Google Scholar 

  16. Scotford IM, Miller PCH (2005) Applications of spectral reflectance techniques in northern European cereal production: a review. Biosystems Eng 90(3):235–250

    Google Scholar 

  17. Govindjee (1995) 63 years since Kautsky—Chlorophyll a fluorescence. Aust J Plant Physiol 22(2):131–160

    Article  CAS  Google Scholar 

  18. Nedbal L, Whitmarsh J (2004) Chlorophyll fluorescence imaging of leaves and fruits. In: Papageorgiu G, Govindjee (eds) Chlorophyll a fluorescence: A signature of photosynthesis. Springer, Dordrecht, The Netherland, pp 389–407

    Google Scholar 

  19. Balachandran S, Osmond CB, Daley PF (1994) Diagnosis of the earliest strain-specific interactions between tobacco-mosaic-virus and chloroplasts of tobacco leaves in-vivo by means of chlorophyll fluorescence imaging. Plant Physiol 104(3):1059–1065

    PubMed  CAS  Google Scholar 

  20. Chaerle L, Hagenbeek D, De Bruyne E, Valcke R, Van der Straeten D (2004) Thermal and chlorophyll-fluorescence imaging distinguish plant-pathogen interactions at an early stage. Plant Cell Physiol 45(7):887–896. doi:10.1093/pcp/pch097

    Article  PubMed  CAS  Google Scholar 

  21. Lazar D, Susila P, Naus J (2006) Early detection of plant stress from changes in distributions of chlorophyll a fluorescence parameters measured with fluorescence imaging. J Fluoresc 16(2):173–176. doi:10.1007/s10895-005-0032-1

    Article  PubMed  CAS  Google Scholar 

  22. Tyystjärvi E, Koski A, Keranen M, Nevalainen O (1999) The Kautsky curve is a built-in barcode. Biophys J 77(2):1159–1167

    Article  PubMed  Google Scholar 

  23. Keranen M, Aro EM, Tyystjarvi E, Nevalainen O (2003) Automatic plant identification with chlorophyll fluorescence fingerprinting. Precis Agric 4(1):53–67. doi:10.1023/A:1021863005378

    Article  Google Scholar 

  24. Bennedsen BS, Rasmussen P, Nielsen H (2000) Crop growth information from line spectrometer data using neural networks. AgEng 2000 Conference, Warwick, UK

    Google Scholar 

  25. Jayas DS, Paliwal J, Visen NS (2000) Multi-layer neural networks for image analysis of agricultural products. J Agric Eng Res 77(2):119–128. doi:10.1006/jaer.2000.0559

    Article  Google Scholar 

  26. Lichtenthaler HK, Buschmann C, Knapp M (2005) How to correctly determine the different chlorophyll fluorescence parameters and the chlorophyll fluorescence decrease ratio RFd of leaves with the PAM fluorometer. Photosynthetica 43(3):379–393. doi:10.1007/s11099-005-0062-6

    Article  CAS  Google Scholar 

  27. Matous K, Benediktyova Z, Berger S, Roitsch T, Nedbal L (2006) Case study of combinatorial imaging: what protocol and what chlorophyll fluorescence image to use when visualizing infection of Arabidopsis thaliana by Pseudomonas syringae?. Photosynth Res 90(3):243–253. doi:10.1007/s11120-006-9120-6

    Article  PubMed  CAS  Google Scholar 

  28. Berger S, Benediktyova Z, Matous K, Bonfig K, Mueller MJ, Nedbal L, Roitsch T (2007) Visualization of dynamics of plant-pathogen interaction by novel combination of chlorophyll fluorescence imaging and statistical analysis: differential effects of virulent and avirulent strains of P. syringae and of oxylipins on A. thaliana. J Exp Bot 58(4):797–806. doi:10.1093/jxb/erl208

    Article  PubMed  CAS  Google Scholar 

  29. Codrea CM, Aittokallio T, Keranen M, Tyystjarvi E, Nevalainen OS (2003) Feature learning with a genetic algorithm for fluorescence fingerprinting of plant species. Pattern Recognit Lett 24(15):2663–2673. doi:10.1016/S0167-8655(03)00109-0

    Article  Google Scholar 

  30. Siedlecki W, Sklansky J (1989) A note on genetic algorithms for large-scale feature selection. Pattern Recognit Lett 10(5):335–347. doi:10.1016/0167-8655(89)90037-8

    Article  Google Scholar 

  31. Nedbal L, Soukupova J, Kaftan D, Whitmarsh J, Trtilek M (2000) Kinetic imaging of chlorophyll fluorescence using modulated light. Photosynth Res 66(1–2):3–12. doi:10.1023/A:1010729821876

    Article  PubMed  CAS  Google Scholar 

  32. Duin RPW (2000) PRTools- version 3.0- A Matlab toolbox for pattern recognition

  33. Fukunaga K (1990) Introduction to statistical pattern recognition, 2nd edn. Academic Press Professional, San Diego, USA

    Google Scholar 

  34. Martinez AM, Kak AC (2001) PCA versus LDA. IEEE Trans Pattern Anal Mach Intell 23(2):228–233. doi:10.1109/34.908974

    Article  Google Scholar 

  35. Wang XC, Paliwal KK (2003) Feature extraction and dimensionality reduction algorithms and their applications in vowel recognition. Pattern Recognit 36(10):2429–2439. doi:10.1016/S0031-3203(03) 00044-X

    Article  Google Scholar 

  36. Bishop CM (1995) Neural networks for pattern recognition. Oxford University Press

  37. Jain AK, Mao JC, Mohiuddin KM (1996) Artificial neural networks: a tutorial. Computer 29(3):31–44. doi:10.1109/2.485891

    Article  Google Scholar 

  38. Fisher R (1936) The use of multiple measurements in taxonomic problems. Ann Eugen 7:179–188

    Google Scholar 

  39. Fukunaga K, Hostetler LD (1975) K-nearest-neighbor Bayes-risk estimation. IEEE Trans Inf Theory 21(3):285–293. doi:10.1109/TIT.1975.1055373

    Article  Google Scholar 

  40. Munne-Bosch S, Nogues S, Alegre L (1999) Diurnal variations of photosynthesis and dew absorption by leaves in two evergreen shrubs growing in Mediterranean field conditions. New Phytol 144(1):109–119. doi:10.1046/j.1469-8137.1999.00490.x

    Article  Google Scholar 

  41. Boser B, Guyon I, Vapnik V (1992) A training algorithm for optimal margin classifiers. COLT 92: Proceedings of the 5th annual workshop on computational learning theory. ACM

  42. Brown M, Gunn SR, Lewis HG (1999) Support vector machines for optimal classification and spectral unmixing. Ecol Modell 120(2–3):167–179. doi:10.1016/S0304-3800(99)00100-3

    Article  Google Scholar 

  43. Camps-Valls G, Gomez-Chova L, Calpe-Maravilla J, Martin-Guerrero JD, Soria-Olivas E, Alonso-Chorda L, Moreno J (2004) Robust support vector method for hyperspectral data classification and knowledge discovery. IEEE Trans Geosci Rem Sens 42(7):1530–1542. doi:10.1109/TGRS.2004.827262

    Article  Google Scholar 

  44. McQueen J (1967) Some methods for classification and analysis of multivariate observations, In Proc. of the 5th Berkeley symposium on mathematical statistics and probability

  45. Jain AK, Duin RPW, Mao JC (2000) Statistical pattern recognition: a review. IEEE Trans Pattern Anal Mach Intell 22(1):4–37. doi:10.1109/34.824819

    Article  Google Scholar 

  46. Kittler J (1978) Feature set search algorithms. Pattern Recognit Signal Process 41–60

  47. Pudil P, Novovicova J, Kittler J (1994) Floating search methods in feature selection. Pattern Recognit Lett 15(11):1119–1125. doi:10.1016/0167-8655(94)90127-9

    Article  Google Scholar 

  48. Somol P, Pudil P, Kittler J (2004) Fast branch & bound algorithms for optimal feature selection. IEEE Trans Pattern Anal Mach Intell 26(7):900–912. doi:10.1109/TPAMI.2004.28

    Article  PubMed  Google Scholar 

  49. Lau KW, Wu QH (2003) Online training of support vector classifier. Pattern Recognit 36(8):1913–1920. doi:10.1016/S0031-3203(03) 00038-4

    Article  Google Scholar 

  50. Chtioui Y, Bertrand D, Barba D (1996) Reduction of the size of the learning data in a probabilistic neural network by hierarchical clustering. Application to the discrimination of seeds by artificial vision. Chemom Intell Lab Syst 35(2):175–186. doi:10.1016/S0169-7439(96)00065-2

    Article  CAS  Google Scholar 

  51. Whitehead D, Walcroft AS, Scott NA, Townsend JA, Trotter CM, Rogers GND (2004) Characteristics of photosynthesis and stomatal conductance in the shrub land species manuka (Leptospermum scoparium) and kanuka (Kunzea ericoides) for the estimation of annual canopy carbon uptake. Tree Physiol 24(7):795–804

    PubMed  CAS  Google Scholar 

Download references

Acknowledgement

The work was supported by the Czech Ministry of Education, Sports and Youth under the Grant MSM6007665808, by the Czech Academy of Sciences Grant AV0Z60870520.

Author information

Authors and Affiliations

  1. Institute of Physical Biology, University of South Bohemia, Zámek 136, 37333, Nové Hrady, Czech Republic

    Anamika Mishra, Karel Matouš, Kumud B. Mishra & Ladislav Nedbal

  2. Institute of Systems Biology and Ecology, Academy of Sciences of the Czech Republic, Zámek 136, 37333, Nové Hrady, Czech Republic

    Anamika Mishra, Karel Matouš, Kumud B. Mishra & Ladislav Nedbal

Authors
  1. Anamika Mishra
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Karel Matouš
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Kumud B. Mishra
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Ladislav Nedbal
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Anamika Mishra.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Mishra, A., Matouš, K., Mishra, K.B. et al. Towards Discrimination of Plant Species by Machine Vision: Advanced Statistical Analysis of Chlorophyll Fluorescence Transients. J Fluoresc 19, 905–913 (2009). https://doi.org/10.1007/s10895-009-0491-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10895-009-0491-x

Keywords

Search

Navigation