The Most Effective Method to Recognize and Assess Bugs in Food Grains

Abstract

Introduction

I.  INTRODUCTION

Because the world's population is growing at a rapid rate and will reach over 9.1 billion by 2050, global food security is a pressing issue (FAO, 2014; Parfitt and others, 2010). According to the World Bank (2011), between 20% and 40% of post-harvest losses (PHL) occur during field and post-harvest operations, with 55% of these losses occurring during storage. According to Asrar, Ashraf, Hussain, Zia, & Rasool (2016), insect infestations worldwide cause damage to food grains that ranges from 10% to 40% annually. According to Jha, Vishwakarma, Ahmad, Rai, & Dixit (2015), cereal storage losses in India ranged from 0.75 to 1.21 percent, while losses in the case of oilseeds and pulses ranged from 1.18 to 1.67 percent and from 0.22 to 1.61 percent, respectively. The number of insects, their stages of development, and the extent of an infestation can be tracked over time through monitoring of stored food grains. It also shows the activity of the insect in relation to the conditions of the environment and determines the effectiveness of insect pest management measures. It is necessary to develop sophisticated insect infestation detection techniques with increased sensitivity in order to lessen these losses and guarantee the secure storage necessary for sustainable agriculture production. This review paper will examine conventional approaches and present a bird's-eye view of recently developed technologies for insect detection in stored products. The acoustic detection of insects in stored food grains has received a lot of attention.

II.  DETECTION OF INSECT PRESENCE

In grain storage facilities, several conventional methods are utilized, the most common of which are visual inspection, probe sampling, and the insect trap method. These strategies are basic yet tedious, work concentrated and emotional. The following sections provide a brief overview of a few well-known strategies. In this study, the workpiece is made of HCHCr-D2 steel. Size is offered in square, flat, and round shapes. This substance is primarily applied in the manufacturing of moulds and dies, and time, pulse off time, and WEDM current affect the rate  of  material  removal  (MMR), as  well  as  the surface roughness (SR). The item code for material HCHCr steel is D2. Table 1 displays the chemical composition of HCHCr-D2 Steel.

Visual inspection can be used to find insects in food grains that have been stored. It is a standard method for comparing quantitative methods because it is uniform, subjective, and qualitative (Semple, 1980). Without drawing grain samples or looking for residual infestation within the storage bags, it is possible to observe the presence of eggs, adult insects, and infested grains with the naked eye. Service of Agribusiness, Fisheries and Food Examiner, England, fostered a few documentations for the utilization of sack.   

(Table 1).

A. Probe Sampling and Trap Method

Sieving and probe sampling are the two most prevalent techniques; some way it is strenuous and monotonous. Tests are utilized in this technique to separate grains (0.5-1 kg) from the put away canister. Sifters are used for screening the bugs from the grains. Grain capacity containers are used to store tests for long periods of time; a regulator genuinely disposes of it and ostensibly explores them, subsequently making it a drawn-out and a portion of the time problematic method These contraptions (see Fig. 2) are helpful for quickly identifying and monitoring insect infestations in stored food grains. The movement of insects toward the air is used as a design idea. The two-in-one probe trap is very effective because it combines a pitfall trap and a probe. It is best suited for capturing pulse beetles because they never stay on the grain surface. A cup with holes and a lid in the shape of a cone is an indicator device. It is held in place at the bottom by a container and a circular dish coated with a sticky substance. The 250 nm ultraviolet light traps embedded with a 4 W germicidal lamp are used in storage at 1.5 meters above the ground. Pheromones are compound substances that bugs emit and are used in bug control traps. These are used for bug-to-bug communication.Fig.1.HCHCR D2 steel

III. OLFACTORY BASED METHODS

A. Solid Phase Micro-Extraction (SPME)

Insect infestation and grain quality evaluation odor detection methods are gaining popularity. Also, this strategy works with early identification of pervasion, stockpiling age assurance, varietal segregation of food grains and so forth. SPME used headspace techniques to separate volatile compounds from vaporized samples. These compounds were then condensed, and gas chromatography-mass spectrometry (GCMS) was used to quantify the volatiles. The SPME method's efficiency and sensitivity are influenced by extraction temperature and time.

B. Electronic Nose (E-nose)

According to Wilson (2012), the intervention of various electronic nose (E-nose) sensor types and instruments is based on the concept of electronic aroma detection (EAD). There are three parts to the e-nose: a system for interpreting the data, a data pre-processor, and a set of smell sensors. The sensor set alters the electrical properties of stored food grains in response to the presence of volatile compounds in the headspace. It includes a predefined database for distinguishing particular volatiles.

In EDM, the tool electrode and workpiece electrode are both completely submerged in a dielectric fluid such as Kerosene oil, EDM grade oil, transformer oil, distilled water, etc. Typically, the tool is made negative by being connected to the cathode (which is called the negative polarity) and made positive by being connected to the anode (which is called the positive polarity). Intermittent electric discharge is used to machine unwanted material because dielectric strength breaks down at high enough voltages.   

Electric discharge machining is done in this way because of the intermittent discharge between the gaps of 10.5 and 125.7 m, which happens after each pulse on duration, produces a very high temperature in a fraction of a second that melts the metal at such a high temperature in the range of 8000oC to 12000oc

IV. ELECTROMAGNETIC-SPECTRUM BASED METHODS

A. Imaging Methods

Machine vision inside noticeable area. An emerging field that combines digital and image processing technology with mechanics, optical instrumentation, electromagnetic sensing, and electromagnetic sensing (Patel et al., 2012). It utilizes the rule of article acknowledgment and characterization based on data extricated from the picture caught by utilizing the camera (Sun, 2016). It is an efficient, consistent, cost-effective, and objective method of inspection that could be used in This non-destructive method helps accelerate the development of automated processes because of its speed and precision, which can meet the ever-increasing demands for production and quality. Image acquisition, image processing or analysis, and recognition and interpretation are the three main steps in machine vision technology. Picture obtaining comprises of catching a genuine picture by utilizing cameras, scanners, recordings and so forth. and converting it into an electronic image.

.

B. X-beam Imaging.

 An encouraging method that makes use of a non-contact sensor to inspect large samples while significantly providing information is X-ray imaging (Yacob et al., 2005). Soft X-ray imaging is a quick, non-destructive, and direct method that is used to find invisible insects in stored food grains (Karunakaran, Jayas, & White, 2003), grade agricultural produce, and determine the internal quality of the produce (Kotwaliwale et al., 2014). The X-ray source that is included in the imaging system; X-beam converter; image capture system and isolated casing (Kotwaliwale, Kalne, & Singh, 2011). A casing separated the imaging medium from the radiation around it and captured the images (Kotwaliwale, Subbiah, Weckler, Brusewitz, & Kranzler, 2007). Electromagnetic rushes of 0.1-10 nm frequencies with 0.12-12 KeV energy are utilized as delicate X-beam for inner examination in exceptionally less time (3-5 s) to create a X-beam picture

Chelladurai, Karuppiah, Jayas, Fields, and White (2014) utilized the delicate X-beam and close infrared (NIR) hyperspectral imaging methods to obtain pictures of soybeans swarmed by C. maculatus alongside uninfested bits. 33 highlights were extricated by delicate X-beam imaging and 48 elements were separated by hyperspectral imaging for information examination. Various phases of invasion were arranged by utilizing LDA and quadratic discriminant investigation (QDA) models. For soft X-ray images, approximately 86% of uninfested and 83% of infested C. maculatus grains were classified by LDA. Principal component analysis (PCA) was used to classify hyperspectral data at wavelengths of 960 nm, 1030 nm, and 1440 nm. The accuracy of the classification of the egg and larval stages was improved by combining X-ray and hyperspectral features.

Thermal imaging. Thermal imaging is  a   night vision

V. NON-IMAGING METHODS

Electronic insect counter with a grain probe (EGPIC). A mechanized detached grain test, known as Electronic Grain Test Bug Counter gains offsite checking and location of bug bothers and remotely shows the information of pervasion levels in put away food grains. A probe, system circuitry, a data logger, and a user interface make up this device. Electric power and circuitry are kept outside the storage structure to avoid the risk of grain dust explosions, and only low voltage, high impedance sensor leads are routed through the grains from the beam generation/detection circuitry to the sensor head. Sensor data is sent to a computer, which analyzes the signals and creates timestamped records of detection.

NIR spectroscopy. Near-infrared spectroscopy, or NIRS, is a method for determining the concentration of biological substances like water, protein, starch, and so on. by measuring the sample's dispersed reflectance, interactance, or transmittance between 780 and 2500 nm. According to Elizabeth, Dowell, Baker, & Throne (2002), it is a non-destructive, speedy, accurate, and cost-effective method that can be used for both internal and external detections in fruits, vegetables, cereals, and pulses. Kim, Phyu, Kim, and Lee, 2003; Xing and Guyer, 2008). Reflectance mode estimates the light reflected or scattered back from the outer layer of the article. When transmittance evaluation makes it difficult to access internal information, interaction mode is acclamatory (Kavdir, Lu, Ariana, & Ngouajio, 2007). Ridgway and Chambers (1996) announced the utilization of NIR reflectance spectroscopy in the recognition of interior pervasion by Sitophilus granarius (L.) in wheat. Chelladurai and others 2014) utilized delicate X-Beam imaging and NIR hyperspectral imaging in the recognition of C. maculatus and characterized the invaded and sound portions of soybean.

A. Acoustic Detection

Acoustic technology is based on the idea that the type and density of insects within a stored grain mass can be estimated by monitoring the sound caused by insects moving and feeding. According to Eliopoulos, Potamitis, Kontodimas, and Givropoulou (2015), it has demonstrated promising results when it comes to the detection of both internal and external insects in the grain mass during the early stages of infestation by means of insect feeding sounds. Mechanical or acoustic waves are detected by acoustic sensors. Acoustic waves are influenced by the properties of the material or object they pass through as well as any obstacles in their path as they travel through it. Because of this, speed or adequacy of the acoustic waves hoses and afterward these progressions are converted into advanced or simple sign through transducers. It for the most part involves the piezoelectric substrate as a sensor. Recognition of covered bugs in grain bit relies upon enhancement and sifting of their development and taking care of sounds. Characterization of designated sounds from different sounds and other restricting elements, for example, sensor responsiveness, sound-commotion proportion, the scope of sensors and so on., restrict the use of acoustic devices. Increased receptivity and credibility have been made possible by technological advancements, including the use of software tools for digital signal processing and improved sensors. When separating the targeted noise from the background noise, pattern features—both spectral and temporal—are also helpful. Insect sounds were separated from background noise using standard speech recognition tools like hidden Markov models and Gaussian mixture models.

Removing grain samples from the storage bin, the insect sound was detected. Above 90% location level could be accomplished at the lower part of the construction. They recommended it as a simple and fast strategy for location and populace thickness assessment of bugs. Kiobia et al. ( 2015) fostered a sound and vibration controlled framework to distinguish Prostephanus shortens and S. Zeamais in maize capacity for dealing with the bug in put away food grains in Sub-Saharan Africa. They discovered that the sensors could measure larval impulses within a 25 cm range. Programming apparatuses were additionally evolved based on contrasts in the phantom and worldly example of bug sounds related with contrasts in physiological exercises. According to Mankin & Moore (2010), feeding sounds had a wider, higher-frequency spectrum than low-energy moving sounds. Trains of sequential bug driving forces (200 ms or less) laid out more huge sign of bug presence than the single motivation (Njoroge et al., 2016, 2017).

Acoustic insect detection probe that operates from the bulk grain's surface to detect and identify the sounds of various species and stages of primary insects in long-term storage. In a 30 kg grain mass, the apparatus can predict live concealed insects of all stages with a confidence factor greater than 90%. Eliopoulos and others 2015) demonstrated that the use of bioacoustics to locate adult wheat beetles within the grain mass was effective. The most significant beetle pests of stored cereals and pulses were used, and the population densities ranged from 1, 2, 10, 20, 50, 100, 200, and 500 beetles per kilogram of grain. The linear model did a great job of explaining the connection between population density and the number of sounds.

B. Signal Processing

Speech recognition techniques are used in the signal processing of recorded sound for the purpose of separating targeted noise from background noise. For the examination of acoustic signs, different sorts of windows are utilized for spectrogram investigation.

C. Filtration and Window Function

A mathematical function used to smoothly draw a sampled signal down to zero at the edges of the sampled region is called a window function (also known as an apodization or tapering function) in signal processing (Prabhu, 2013). The window function's spectrum can be given in both the frequency and time domains. More than one sample (at least two full cycles) is used in acoustic signal analysis.

D. Acoustic Range Highlights

Segregation of sign clamor should be possible by isolating particular sound created by the bugs by registering specific highlights of every motivation and afterward they are contrasted and the standardized otherworldly elements (Potamitis et al., 2009). It was reported that the following spectral features could be used for discrimination: The dominant harmonic, linear frequency spectrum coefficients, and the Fourier transform (Mankin et al., 2009). In order to distinguish the features of background noise, such as clustering, insect-produced sound has been used. vector quantization and Gaussian combination models (Potamitis et al., 2009).

VI. ACOUSTIC WORLDLY EXAMPLE HIGHLIGHTS.

 By consolidating temporal pattern features, insect detection accuracy can be improved. Many insects have predictable behavioral patterns. A portion of the examples with inadequate normalities were not described dependably by the PC programs, yet a few examinations observed that these examples were recognized as an explosion of motivations isolated by calm time frames s or more (Mankin et al., 2009). Insect feeding or movement resulted in a burst of approximately six impulses. To distinguish a specific insect sound from the background sounds, this phenomenon can be used as signal features. ID of transient highlights is simple for enormous and dynamic bugs. Union of blasts as a sign handling highlight empowers the evacuation of wind-prompted catching clamor or other foundation commotion like the sound signs delivered by bugs

Table-2

                                   Advantages and limitations of modern methods of insect detection in stored food grains.

Acoustic signals in biological materials are typically analyzed using gaussian windows (Mannell, 2008). Prior to processing, complex acoustic signals are filtered using a low-pass (LP), high-pass (HP), and band pass (BP) filter. During separating, a few frequencies are permitted to pass and others ought to be hindered. The majority of filters have a zone around the cut-off frequency where some frequencies can pass through.

According to Mannell (2008), this facilitates a smoother transition between the pass-band, or unattenuated frequencies, and the stop-band, or attenuated frequencies.

The term of R. ferrugineus sound motivation was 3-50 ms and 3.8 kHz top recurrence range (Potamitis et al., 2009). According to Mankin & Moore (2010), background noise typically had a high relative energy peak at low frequencies (below 1 kHz). The nature and degree of the background noise determine the filtering requirements. Signals with amplitudes above the threshold were eliminated following filtering. The sensor placements had an effect on the frequency and amplitude of insects. Before proceeding with signal processing, signals from one sensor can be subtracted from those from the other if multiple sensors are utilized.

A. Acoustic Range Highlights

Segregation of sign clamor should be possible by isolating particular sound created by the bugs by registering specific highlights of every motivation and afterward they are contrasted and the standardized otherworldly elements (Potamitis et al., 2009). It was reported that the following spectral features could be used for discrimination: The dominant harmonic, linear frequency spectrum coefficients, and the Fourier transform (Mankin et al., 2009). In order to distinguish the features of background noise, such as clustering, insect-produced sound has been used. vector quantization and Gaussian combination models (Potamitis et al., 2009).

B. Acoustic Worldly Example Highlights

By consolidating temporal pattern features, insect detection accuracy can be improved. Many insects have predictable behavioral patterns. A portion of the examples with inadequate normalities were not described dependably by the PC programs, yet a few examinations observed that these examples were recognized as an explosion of motivations isolated by calm time frames s or more (Mankin et al., 2009). Insect feeding or movement resulted in a burst of approximately six impulses. To distinguish a specific insect sound from the background sounds, this phenomenon can be used as signal features. ID of transient highlights is simple for enormous and dynamic bugs. As a signal processing feature, consolidation of bursts makes it possible to get rid of wind-induced trapping noise or other background noise that looks like insect sound signals.

Conclusion

There are a number of options for identifying an infestation of insects in stored food grains. Visual inspection is one of the conventional methods; however, it is not suitable for bulk storage or the detection of hidden or low-density infestations; besides taking a long time. While sampling probes and traps can be useful, they are time-consuming, tedious, only provide temporal data, and sometimes necessitate the destruction of samples. The visual lure is a chemical-free method that uses light to attract insects, but its accuracy is influenced by the environment, has low sensitivity, and only provides information about adult insects. The pheromones method does not require sampling and can detect both internal and external insects. However, due to the lack of information on various pheromone characteristics, its detection efficacy is limited to a small number of insect species. Berlese pipe strategy is quite possibly of the most well-known technique utilized in grain lifts, yet it is slow, unfit to identify stowed away pervasion and exactness relies upon bugs\' size. In many developing nations, the uric acid detection method is officially accepted, but it only works on highly infested grains and has low sensitivity. The hidden infestation detector is a straightforward and inexpensive device that can only be used in very small quantities and is destructive. The majority of conventional methods are only effective for external detection. The electrical conductance method, which was recently developed, can find the hidden infestation, but it can\'t find the egg and larval stages, needs a lot of moisture, and can only be used to characterize a single grain at a time. Although the solid phase micro extraction method with dynamic headspace has a high sensitivity, it can only detect adult stages and necessitates the expertise of a skilled individual for operation and analysis. The fast, objective, and poison-free high technology of an electronic nose for insect detection is limited by the need for a large amount of experimental data to train the sensors; other than the sensors are impacted by the natural elements and furthermore their productivity diminishes over the long run. Machine Vision framework working in apparent reach is utilized for different applications in farming tasks and it shows the capacity to group the grains however distinguishing the dead and interior insects is costly and unfit.

References

[1] W. Laopongsit, G. Srzednicki, and J. Craske A preliminary investigation into the use of solid phase micro-extraction (SPME) to identify insect infestations in wheat grain 59, 88–95, Journal of Stored Products Research. [2] M. P. Leblnac, D. Gaunt, and F. Fleurat-Lessard are the authors. Evaluation of the potential of acoustic equipment for predicting infestation risk over extended storage periods—an experimental study of its use for real-time insect detection in grain bins. IOBC-WPRS Notice, 69, 79-88. [3] Litzkow, C. A., Shuman, D., Kruss, S., and Coffelt, J. A. (1997). Patent Number: 5,646,404. DC, Washington: United States Patent and Trademark Office. [4] Magan, N., and Evans, P. (2000). Volatiles as a sign of parasitic movement and separation among species, and the expected utilization of electronic nose innovation for early location of grain deterioration. Diary of Put away Items Exploration, 36(4), 319-340. [5] E. B. Maghirang, F. E. Dowell, J. E. Baker, and J. E. Throne Utilizing near-infrared reflectance spectroscopy, automated detection of single wheat kernels containing either live or dead insects. ASAE Transactions, 46(4), 1277. [6] Manickavasagan, A., Jayas, D. S., and White, N. D. G. (2008). Thermal imaging to locate Cryptolestes ferrugineus infestations within wheat kernels. Diary of Put away Items Exploration, 44(2), 186-192. [7] R. Mankin and D. W. Hagstrum are the authors. Insects are being monitored by sound. Protection of Stored Products, 263 [8] R. W. Mankin, A. Moore, and others Acoustic oryctes rhinoceros detection (Coleoptera: Scarabaeidae: Dynastinae) and nasuti termes luzonicus (Isoptera: Termitidae) in urban Guam\'s palm trees. Diary of Financial Entomology, 103(4), 1135-1143. Mankin, R. W., Samson, P. R., and Chandler, K. J. (2009). [9] [9] Melolonthine larvae were detected by sound in Australian sugarcane. 102(4):1523–1535, Journal of Economic Entomology. [10] R. Mannell (2008) Acoustics of speech. analog_digital.html can be found at http://clas.mq.edu.au/speech/acoustics/frequency. [11] J. M. Minkevich, C. J. Demianyk, N. D. G. White, D. S. Jayas, B. Timlick, and others A quick technique to distinguish Cryptolestes ferrugineus (Coleoptera: larvae in stored grain (Cucujidae). Canadian Diary of Plant Science, 82(3), 591-597. [12] S. Mohan (2007) Post-harvest eco-friendly technologies for managing stored grain insects. 1, 45–47, Green Farming. Sreenarayanan, V. V., Mohan, S., M. Gopalan, P. C. Sundarababu, and Practical studies on the management of Rhyzopertha Dominica (F.) in rice warehouses using light traps and bait traps. 40(2), 148–152, International Journal of Pest Management. [13] Nanje Gowda, N. A., and Alagusundaram, K. (2013). Utilization of warm imaging to further develop the food grains quality during stockpiling. Internat. Current Agriculture Res, 1(7), 34–41. [14] Narendra, V. G., and Hareesh, K. S. (2010). A review of the prospects for quality evaluation of agricultural and food products using automated grading and sorting systems based on computer vision. 2(1), 43–65, International Journal of Computer Application. [15] White, N. D. G., S. Neethirajan, C. Karunakaran, D. S. Jayas, and Identification methods for put away item bugs in grain. Food Control, 18(2), 157-162. [16] Hardy, G. Niu, M. Agarwal, L. Hua, Y. Ren, and others Utilizing solid phase micro extraction–gas chromatography mass spectrometry (SPME–GCMS), the volatiles of Tribolium castaneum (H.) in flour were characterized. Food Science and Human Wellbeing, 5(1), 24-29. [17] A. W. Njoroge, H. Affognon, C. Mutungi, U. Richter, O. Hensel, and B. Rohde are among the authors. (2017). Acanthoscelides obtectus\' bioacoustics (Coleoptera: Chrysomelidae

Copyright

Copyright © 2023 Gowsika M, Subiksha I, Antony Joseph Raj D, Sudharsanavel V, Annapoorani V. 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.