Near infrared (NIR) hyperspectral imaging to classify fungal infected date fruits

Teena, M.; Manickavasagan, Annamalai; Ravikanth, Lankapalli; Jayas, Digvir S.

doi:10.1016/j.jspr.2014.09.005

Cited by 65 publications

(30 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, in the classifications of different stages of infection of Khalas dates, the highest classification accuracies reported by both LDA and QDA were 100%; which was better than the results discussed in Teena et al (2014) (LDA-97%, QDA-100%) for the same date variety using NIR hyperspectral imaging. From these results, it can be inferred that color imaging could provide equal or better classification than hyperspectral imaging to discriminate the infected samples of Khalas variety.…”

Section: Six Class Modelmentioning

confidence: 49%

“…For a uniform distribution of spores in the collected spore suspension, a drop of Tween 20 was added and mixed thoroughly. This spore suspension was examined for its spore concentration using a hemocytometer (Neubauer, China) and was found to have 107 spores/ml (Rahman et al, 2004;Teena et al, 2014).…”

Section: Artificial Inoculation Of Dates With a Flavusmentioning

confidence: 99%

“…The PROC DISCRIM procedure was performed in SAS to discriminate control dates from fungus infected dates of Fard, Khalas and Naghal using the 64 extracted features. Both LDA and QDA with leave-one-out crossvalidation method was used for developing a two-class model (between each control and IS (five infection stages combined)), six-class model (between each control and IS (five infection stages in separate classes) and pair-wise model (between each control and each infection stages of IS as pairs) (Teena et al, 2014).…”

Section: Classification Using 64 Featuresmentioning

confidence: 99%

“…Early detection of fungal infection in dates would be highly beneficial in taking corrective action plan (Teena et al, 2014). Despite several studies on date quality using nondestructive computer vision techniques, at present there is no published work on the detection of fungal infection using color imaging (Teena et al, 2013).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

RGB color imaging to detect Aspergillus flavus infection in dates

Teena

Manickavasagan

Al-Sadi³

et al. 2016

Emir. J. Food Agric

View full text Add to dashboard Cite

Section: Six Class Modelmentioning

confidence: 49%

Section: Artificial Inoculation Of Dates With a Flavusmentioning

confidence: 99%

Section: Classification Using 64 Featuresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

RGB color imaging to detect Aspergillus flavus infection in dates

Teena

Manickavasagan

Al-Sadi³

et al. 2016

Emir. J. Food Agric

View full text Add to dashboard Cite

“…To solve the problem, a near-infrared spectroscopy and artificial neural networks have been used to predict the hardness and the other quality parameters of wheat [6][7][8], although the corresponding measurement accuracy is easily affected by the moisture of kernels. In the recent years, acoustic methods have been put forward.…”

Section: Introductionmentioning

confidence: 99%

Detection of wheat hardness based on a laser-generated ultrasonic signal

Fan¹

2017

Ukr. J. Phys. Opt.

View full text Add to dashboard Cite

Abstract.Hardness is an important index for wheat quality, which determines its usage, price and wheat-processing techniques. Therefore accurate measurements of the hardness of kernels is a key problem for assessing wheat quality. In this work we develop a new method for testing the hardness basing on a laser-induced ultrasonic signal. We describe the measurement principles, as well as sampling and preprocessing of ultrasonic signals. The acoustic signal is analyzed both in the time and frequency domains, using fast Fourier, discrete cosine (DCT) and wavelet (WT) transforms. The main eight parameters whose correlation indices surpass the threshold of 0.8 are selected as feature characteristic parameters of the hardness. They include a waveform index T6, a pulse factor T7, sub-band energy ratios SER 1 -SER 3 , a sum of the DCT magnitudes, and two wavelet parameters, WTF1 and WTF2. A testing model of the hardness is built basing on a standard extreme learning machine (ELM) algorithm and using all of these feature parameters as an input. A number of experiments have been performed on twenty wheat varieties with different hardness indices. The results show that the maximal relative measurement error and the mean relative error are approximately equal to -3% and 1%, respectively. As a result, our method of measuring the wheat hardness, which is based on combination of laser ultrasonic waves and ELM, is feasible and accurate enough.

show abstract

Detection and identification of fungal growth on freeze‐dried Agaricus bisporus using spectra and olfactory sensors

Wang

Pei

et al. 2020

J Sci Food Agric

View full text Add to dashboard Cite

BACKGROUND Fungal contamination in food products leads to mustiness, biochemical changes, and undesirable odors, which result in lower food quality and lower market value. To develop a rapid method for detecting fungi, hyperspectral imaging (HSI) was applied to identify five fungi inoculated on plates (Aspergillus niger, Aspergillus flavus, Penicillium chrysogenum, Aspergillus fumigatus, and Aspergillus ochraceus). Near‐infrared (NIR) spectroscopy, mid‐infrared (MIR) spectroscopy, and an electronic nose (E‐nose) were applied to detect and identify freeze‐dried Agaricus bisporus infected with the five fungi. RESULTS Partial least squares regression (PLSR) models were used to distinguish the HSI spectra of the five fungi on the plates. The A. ochraceus group had the highest calibration performance: coefficient of calibration (Rc2) = 0.786, root mean‐square error of calibration (RMSEC) = 0.125 log CFU g−1. The A. flavus group had the highest prediction performance: coefficient of prediction (Rp2) = 0.821, root mean‐square error of prediction (RMSEP) = 0.083 log CFU g−1. The ratio of performance deviation (RPD) values of all of the models was higher than 2.0 for the NIR, MIR, and E‐nose results for freeze‐dried A. bisporus infected with different fungi. The fungal species and degree of infection can be distinguished by principal component analysis (PCA) and partial least squares discriminant analysis (PLS‐DA) using NIR, MIR, and E‐nose, as the discrimination accuracy was more than 90%. The NIR methods had a higher recognition rate than the MIR and E‐nose methods. CONCLUSION Principal component analysis (PCA) and PLSR models based on full spectra of HSI can achieve good discrimination results for these five fungi on plates. Moreover, NIR, MIR, and the E‐nose were proven to be effective in monitoring fungal contamination on freeze‐dried A. bisporus. However, NIR could be a more accurate method. © 2020 Society of Chemical Industry

show abstract

Near infrared (NIR) hyperspectral imaging to classify fungal infected date fruits

Cited by 65 publications

References 30 publications

RGB color imaging to detect Aspergillus flavus infection in dates

RGB color imaging to detect Aspergillus flavus infection in dates

Detection of wheat hardness based on a laser-generated ultrasonic signal

Detection and identification of fungal growth on freeze‐dried Agaricus bisporus using spectra and olfactory sensors

Contact Info

Product

Resources

About