Green Fruit Detection with a Small Dataset under a Similar Color Background Based on the Improved YOLOv5-AT.

Green fruit detection is of great significance for estimating orchard yield and the allocation of water and fertilizer. However, due to the similar colors of green fruit and the background of images, the complexity of backgrounds and the difficulty in collecting green fruit datasets, there is currently no accurate and convenient green fruit detection method available for small datasets. The YOLO object detection model, a representative of the single-stage detection framework, has the advantages of a flexible structure, fast inference speed and excellent versatility. In this study, we proposed a model based on the improved YOLOv5 model that combined data augmentation methods to detect green fruit in a small dataset with a background of similar color. In the improved YOLOv5 model (YOLOv5-AT), a Conv-AT block and SA and CA blocks were designed to construct feature information from different perspectives and improve the accuracy by conveying local key information to the deeper layer. The proposed method was applied to green oranges, green tomatoes and green persimmons, and the mAPs were higher than those of other YOLO object detection models, reaching 84.6%, 98.0% and 85.1%, respectively. Furthermore, taking green oranges as an example, a mAP of 82.2% was obtained on the basis of retaining 50% of the original dataset (163 images), which was only 2.4% lower than that obtained when using 100% of the dataset (326 images) for training. Thus, the YOLOv5-AT model combined with data augmentation methods can effectively achieve accurate detection in small green fruit datasets under a similar color background. These research results could provide supportive data for improving the efficiency of agricultural production.

Long-range infrared absorption spectroscopy and fast mass spectrometry for rapid online measurements of volatile organic compounds from black tea fermentation.

Fermentation is the key process to determine the quality of black tea. Traditional physical and chemical analyses are time consuming, it cannot meet the needs of online monitoring. The existing rapid testing techniques cannot determine the specific volatile organic compounds (VOCs) produced at different stages of fermentation, resulting in poor model transferability; therefore, the current degree of black tea fermentation mainly relies on the sensory judgment of tea makers. This study used proton transfer reaction mass spectrometry (PTR-MS) and fourier transform infrared spectroscopy (FTIR) combined with different injection methods to collect VOCs of the samples, the rule of change of specific VOCs was clarified, and the extreme learning machine (ELM) model was established after principal component analysis (PCA), the prediction accuracy reached 95% and 100%, respectively. Finally, different application scenarios of the two technologies in the actual production of black tea are discussed based on their respective advantages.

Gas Imaging with Uncooled Thermal Imager.

Gas imaging has become one of the research hotspots in the field of gas detection due to its significant advantages, such as high efficiency, large range, and dynamic visualization. It is widely used in industries such as natural gas transportation, chemical, and electric power industries. With the development of infrared detector technology, uncooled thermal imagers are undergoing a developmental stage of technological advancement and widespread application. This article introduces a gas imaging principle and radiation transfer model, focusing on passive imaging technology and active imaging technology. Combined with the actual analysis, the application scenarios using uncooled thermal imaging cameras for gas imaging measurement are analyzed. Finally, the limitations and challenges of the development of gas imaging technology are analyzed.

Rapid Determination of Different Ripening Stages of Occidental Pears (Pyrus communis L.) by Volatile Organic Compounds Using Proton-Transfer-Reaction Mass Spectrometry (PTR-MS).

Determination of Occidental pear (Pyrus communis) ripening is difficult because the appearance of Occidental pears does not change significantly during the ripening process. Occidental pears at different ripening stages release different volatile organic compounds (VOCs), which can be used to determine fruit ripeness non-destructively and rapidly. In this study, VOCs were detected using proton-transfer-reaction mass spectrometry (PTR-MS). Notably, data were acquired within 1 min. Occidental pears harvested at five separate times were divided into three ripening stages: unripe, ripe, and overripe. The results showed that the composition of VOCs differed depending on the ripening stage. In particular, the concentrations of esters and terpenes significantly increased during the overripe stage. Three ripening stages were clearly discriminated by heatmap clustering and principal component analysis (PCA). This study provided a rapid and non-destructive method to evaluate the ripening stages of Occidental pears. The result can help fruit farmers to decide the optimum harvest time and hence reduce their economic losses.

Non-destructive and in-situ detection of shrimp freshness using mid-infrared fiber-optic evanescent wave spectroscopy.

There is strong interest in non-destructive and rapid determination of food freshness in food research. In this study, mid-infrared (MIR) fiber-optic evanescent wave (FOEW) spectroscopy was applied to monitor shrimp freshness through the evaluation of protein, chitin, and calcite contents in conjunction with a Partial Least Squares Discriminant Analysis (PLS-DA) model. Shrimp shells were wiped with a micro fiber-optic probe to obtain a FOEW spectrum which quickly and nondestructively allowed evaluation of the shrimp freshness. Peaks for proteins, chitin, and calcite, which are closely related to shrimp freshness, were detected and quantified. Compared with the standard indicator for evaluating shrimp freshness (total volatile basic nitrogen), the PLS-DA model gave recognition rates for shrimp freshness using calibration and validation sets of the FOEW data of 87.27%, 90.28%, respectively. Our results show that FOEW spectroscopy is a feasible method for non-destructive and in-site detection of shrimp freshness.

Rapid identification of artificial fragrant rice based on volatile organic compounds: From PTR-MS to FTIR.

The volatile organic compounds (VOCs) released from foods can reflect their internal properties. Artificial fragrant rice (AFR) is a fraudulent food product in which the flavor of low-quality rice is artificially enhanced by addition of essence. In this study, proton-transfer reaction mass spectrometry, long optical path gas phase FTIR spectroscopy and fiber optic evanescent wave were used to analyze the characteristic mass-charge ratios signal and infrared fingerprint signal of four essence which may be used to make AFR, and the prepared AFR samples with different essence levels (0.001 %-0.3 %) were used to verify the detection performance of the detection methods. The results show that the three detection methods effectively identified AFR containing the minimum recommended dose of essence (≥0.1 %, w/w). The above detection methods can provide detection results in real time without complex sample pretreatment and provide options as rapid screening methods for food regulatory authorities to identify AFR.

A Real-Time Detection Method of Hg2+ in Drinking Water via Portable Biosensor: Using a Smartphone as a Low-Cost Micro-Spectrometer to Read the Colorimetric Signals.

This paper reported a real-time detection strategy for Hg2+ inspired by the visible spectrophotometer that used a smartphone as a low-cost micro-spectrometer. In combination with the smartphone's camera and optical accessories, the phone's built-in software can process the received light band image and then read out the spectral data in real time. The sensor was also used to detect gold nanoparticles with an LOD of 0.14 µM, which are widely used in colorimetric biosensors. Ultimately, a gold nanoparticles-glutathione (AuNPs-GSH) conjugate was used as a probe to detect Hg2+ in water with an LOD of 1.2 nM and was applied successfully to natural mineral water, pure water, tap water, and river water samples.

In situ detection of fruit spoilage based on volatile compounds using the mid-infrared fiber-optic evanescent wave spectroscopy.

Volatile compounds such as ethanol released from fruit can be rapidly detected using Fourier Transform Infrared spectroscopy based on a long-path gas cell. However, this method relies on a long optical path length and requires pumping fruit volatiles into the gas cell. This can lead to the volatile compounds being contaminated and not detectable in situ. Fiber optic evanescent wave spectroscopy (FOEW) is not influenced by the path length so can detect materials (solid, liquid and gas phase) rapidly in situ, using only a few millimeters of optical fiber. In the present study, a spiral silver halide FOEW sensor with a length of approximately 21 mm was used to replace a long-path gas cell to explore the feasibility of identifying volatile compounds released from grapes in situ. The absorption peaks of ethanol in the volatile compounds were clearly found in the FOEW spectra and their intensity gradually increased as the storage time of the grapes increased. PCA analysis of these spectra showed clear clustering at different storage times (1-3, 4-5 and 6-7 d), revealing that the concentration of the ethanol released from the grapes changed significantly with time. The qualitative model established by PLS-DA algorithm could accurately classify grape samples as "Fresh," "Slight spoilage," or "Severe spoilage". The accuracy of the calibration and validation sets both were 100.00%. These changes can therefore be used for rapidly identifying fruit deterioration. Compared with the method used in a previous study by the authors, this method avoids using a pumping process and can thus identify volatile compounds and hence monitor deterioration in situ and on-line by placing a very short optical fiber near the fruit.

Rapid quantitative analysis of potassium in soil based on direct-focused laser ablation-laser induced breakdown spectroscopy.

A fast quantitative analysis method of soil potassium based on direct-focused laser ablation-laser induced breakdown spectroscopy (direct-focused LA-LIBS) was proposed and tested. A high single-pulse energy laser (200 mJ/pulse) beam was focused on the aerosols near the focus of the 10 kHz fiber laser to generate plasma spectra, and the analytical capability of the direct-focused LA-LIBS system was compared with traditional LIBS system using a high single-pulse energy laser (SP-LIBS). The result showed that for moist soil samples the data stability of the direct-focused LA-LIBS method was significantly improved and the R2 factor of the calibration curve improved from 0.64 to 0.93, the limit of detection improved from 159.2 µg/g to 140.9 µg/g. Three random soil samples from different areas of Beijing suburbs were analyzed by the direct-focused LA-LIBS method, and the results were consistent with AAS. The direct-focused LA-LIBS method proposed is different from the traditional double-pulse technology and laser ablation-assisted technology because it not only does not need carrier gas, but also can overcome the matrix differences better, especially the influence of moisture, which provides a new idea for the rapid detection of nutrient elements in field soils.

Visualizing changes of metabolites during iron deficiency chlorosis in field-grown pear leaves using micro-Raman spectral imaging.

Owing to iron chlorosis, pear trees are some of the most severely impacted by iron deficiency, and they suffer significant losses every year. While it is possible to determine the iron content of leaves using laboratory-standard analytical techniques, the sampling and analysis process is time-consuming and labor-intensive, and it does not quickly and accurately identify the physiological state of iron-deficient leaves. Therefore, it is crucial to find a precise and quick visualization approach for metabolites linked to leaf iron to comprehend the mechanism of iron deficiency and create management strategies for pear-tree planting. In this paper, we propose a micro-Raman spectral imaging method for non-destructive, rapid, and precise visual characterization of iron-deficiency-related metabolites in pear leaves. According to our findings, iron deficiency significantly decreased the Raman peak intensities of chlorophylls and lipids in leaves. The spatial distributions of chlorophylls and lipids in the leaves changed significantly as the symptoms of iron insufficiency worsened. The technique offers a new, prospective tool for rapid recognition of iron deficiency in pear trees because it is capable of visual detection of plant physiological metabolites induced by iron deficiency.

Rapid determination of trace cadmium in drinking water using laser-induced breakdown spectroscopy coupled with chelating resin enrichment.

The determination of heavy metals in drinking water is of great importance, but it is hard to realize rapid and in-situ measurement. Laser-induced breakdown spectroscopy is an effective method for both solid and liquid sample analysis with advantages of fast and micro-destructive. However, the concentrations of heavy metals in drinking water is too low to be directly detected using LIBS. In this study, we enhanced the sensitivity of LIBS by coupling with chelating resin, which is usually used for water purification. The resin provided a rapid enrichment of the heavy metal, so the limits of detection of common LIBS system was much enhanced. Using Cadmium as the representative heavy metal, PLSR model for predicting Cd were built based on the spectral intensity (Cd 214.4 nm) with concentrations from 0 to 100 µg/L, and resulted in correlation coefficient of 0.94433 and RMSE of 7.1517 µg/L. The LoD was 3.6 µg/L. Furthermore, the volume, resin mass, adsorption time, and LIBS system parameters were optimized for practical applications. We also demonstrated that the resin can be recycled without loss in sensing ability. The combination of chelating resin with LIBS provides inexpensive, rapid, and sensitive detection method of trace heavy metal contaminants in drinking water.

Detecting volatile compounds in food by open-path Fourier-transform infrared spectroscopy.

We previously found that the brand of a food and spoilage of the food can be identified from the infrared spectra of the volatile compounds released. However, this required pumping the volatile compounds into a gas cell, meaning measurements over large areas could not be made. Gas components can be quantified from a distance of a few metres or kilometres by open-path Fourier-transform infrared spectroscopy (FTIR), and the spatial distributions of gas clouds can even be determined using open-path FTIR and an imaging detection method. In the study described here, we used open-path FTIR to remotely detect volatile compounds in food. Active and passive methods were used to obtain infrared spectra of volatile compounds released from spirits, vinegars, and grapes from a distance of 5 m. The absorption characteristics of ethanol, esters, and unknown volatile compounds were clearly found in the spectra. The brands of the spirits and degree to which the grapes had spoiled were identified by compensating for ethanol in the atmosphere and chemometrics. The results indicate that open-path FTIR can be used to remotely detect volatile compounds released by food and may be able to be used to identify spoiling food in large food warehouses.

Potential using of infrared thermal imaging to detect volatile compounds released from decayed grapes.

Previous studies have demonstrated variations in volatile compound content during fruit spoilage. Infrared spectroscopy was proposed as an alternative method to discriminate the various states of decayed fruit through the makeup of their volatile compounds. Based on the infrared spectra of volatile compounds obtained from decayed grapes, this study simplified the extraction of their feature spectra and visualized their gas plumes by using a commercial infrared thermal camera equipped with a custom-made wavelength filter. As a function of volatilization gradients, accumulated gray value and imaging area were proposed as indicators for semi-quantitative analysis in a volatilization range similar to that of ethanol solutions ranging from 10% to 70%. Fresh, seriously decayed, and slightly or moderately decayed grapes were rapidly discriminated through their alcoholic volatiles by thermal images with correct classification ratings of 100%, 93.3%, and 90%, respectively.

Ultrasensitive nanoparticle enhanced laser-induced breakdown spectroscopy using a super-hydrophobic substrate coupled with magnetic confinement.

In this study, we developed a substrate to enhance the sensitivity of LIBS by 5 orders of magnitude. Using a combination of field enhancement due to the metal nanoparticles in the substrate, the aggregate effect of super-hydrophobic interfaces and magnetic confinement, we performed a quantitative measurement of copper in solution with concentrations on the ppt level. We also demonstrated that the substrate improves quantitative measurements by providing an opportunity for internal standardization.

Compensation method for the influence of angle of view on animal temperature measurement using thermal imaging camera combined with depth image.

In the study, we proposed an animal surface temperature measurement method based on Kinect sensor and infrared thermal imager to facilitate the screening of animals with febrile diseases. Due to random motion and small surface temperature variation of animals, the influence of the angle of view on temperature measurement is significant. The method proposed in the present study could compensate the temperature measurement error caused by the angle of view. Firstly, we analyzed the relationship between measured temperature and angle of view and established the mathematical model for compensating the influence of the angle of view with the correlation coefficient above 0.99. Secondly, the fusion method of depth and infrared thermal images was established for synchronous image capture with Kinect sensor and infrared thermal imager and the angle of view of each pixel was calculated. According to experimental results, without compensation treatment, the temperature image measured in the angle of view of 74° to 76° showed the difference of more than 2°C compared with that measured in the angle of view of 0°. However, after compensation treatment, the temperature difference range was only 0.03-1.2°C. This method is applicable for real-time compensation of errors caused by the angle of view during the temperature measurement process with the infrared thermal imager.

An instantaneous approach for determining the infrared emissivity of swine surface and the influencing factors.

Infrared thermal imaging technology has been widely employed in temperature measurements of human and animals and its accuracy relies on the determination process of the emissivity of the target to a large extent. However, common used methods were unable to determine the emissivity of the surface of living animals and thus lower the accuracy. In this paper, we suggested a new approach to acquire the infrared emissivity of living swine in real time. In the approach, the surface temperature of swine and reference body were measured to compute the emissivity and the measurement process was completed in a non-contact and non-invasive manner. We changed the surface reflection energy of animals and reference body by changing the ambient radiant energy and obtain the surface emissivity in real time without confirming the actual temperature of animal surface. In this way, the infrared emissivity of the animal surface can be determined instantaneously and without knowing the real temperature. Both swine specimen and a living swine were used in this study. Using this method, we measured the emissivity of different body sites of the swine. The results showed that the emissivity values at different body sites show the significant differences. The emissivity values at trotter and eye were respectively 0.895 and 0.930 and the emissivity on swine surface varied from 0.945 to 0.978. More important, the distribution of the infrared emissivity on a living swine was explored and the detailed differences of the emissivity on a swine surface can be cleanly seen. Furthermore, we studied the influencing factors on the emissivity of animal surface, through measuring the emissivity distribution on swine surface when pig specimens were sprayed with water on the surface or heated using this method. This study is of great significance for the accurate measurement of swine surface temperature.

[Research on Grape Deterioration Process via Volatiles -Using Long Optical-Path Infrared Spectroscopy and Simplified E-Nose].

Grapes vaporize volatiles in specific compositions and concentrations during deterioration processes. Our previous study demonstrated that it is possible to analyze grapes spoilage stages by using the infrared spectra of their volatiles. However, only the spectral characteristics of alcohol, ethyl acetate and carbon dioxide were observed in the experiment because of the low concentration of the volatiles. In this paper, the sensitivity of the spectrometry system was enhanced by increasing the optical-path with multi-reflecting mirrors. We used the new spectrometry system to study the details of the infrared spectra of the volatiles from grapes during spoilage, and observed the spectral characteristics of several kinds of ethanol, esters, aldehyde and ethylene. The concentrations of some components in the volatiles changes with storage time, which can be a biomarker to represent the spoilage stages of grapes. Chemometrics were used to analyze the spectral bands of ethanol and esters, demonstrating there are obvious differences between fresh and decayed grapes. Furthermore, we developed a simplified E-nose system comprised by sensor array, based on the results of spectral analysis. The classification and discrimination of grape spoilage were tested with E-nose. This was a further study of the previous publication and had given a more precise observation of the infrared spectral characteristics of the volatiles from decayed grapes. This study provided a basis for developing real-time monitoring techniques of fruits deterioration.

[Application of Fourier transform infrared spectroscopy in identification of wine spoilage].

In the present work, fresh and spoiled wine samples from three wines produced by different companies were studied u- sing Fourier transform infrared (FTIR) spectroscopy. We analyzed the physicochemical property change in the process of spoil- age, and then, gave out the attribution of some main FTIR absorption peaks. A novel determination method was explored based on the comparisons of some absorbance ratios at different wavebands although the absorbance ratios in this method were relative. Through the compare of the wine spectra before and after spoiled, the authors found that they were informative at the bands of 3,020~2,790, 1,760~1,620 and 1,550~800 cm(-1). In order to find the relation between these informative spectral bands and the wine deterioration and achieve the discriminant analysis, chemometrics methods were introduced. Principal compounds analysis (PCA) and soft independent modeling of class analogy (SIMCA) were used for classifying different-quality wines. And partial least squares discriminant analysis (PLS-DA) was applied to identify spoiled wines and good wines. Results showed that FTIR technique combined with chemometrics methods could effectively distinguish spoiled wines from fresh samples. The effect of classification at the wave band of 1 550-800 cm(-1) was the best. The recognition rate of SIMCA and PLSDA were respectively 94% and 100%. This study demonstrates that Fourier transform infrared spectroscopy is an effective tool for monitoring red wine's spoilage and provides theoretical support for developing early-warning equipments.

Analyzing strawberry spoilage via its volatile compounds using longpath Fourier transform infrared spectroscopy.

The volatile compounds from fruits vary based on the spoilage stage. We used FTIR spectroscopy to analyze and to attempt to identify the spoilage process of strawberries. To enhance the sensitivity of the measuring system, we increased the optical pathlength by using multi-reflecting mirrors. The volatile compounds that were vaporized from strawberries in different spoilage stages were tested. We analyzed the spectra and found that the concentrations of esters, alcohols, ethylene, and similar compounds changed with deterioration. The change patterns of the infrared spectra for the volatiles were further examined using 2D correlation spectroscopy. We analyzed the spectral data using PCA and were able to distinguish the fresh, slightly spoiled strawberries from the seriously spoiled strawberries. This study demonstrates that FTIR is an effective tool for monitoring strawberry spoilage and for providing status alerts.

[Pretreatment method of near-infrared diffuse reflection spectra used for sugar content prediction of pears].

The content of sugar is an important quality index for pears. However, the traditional sugar measurement methods are time-consuming and destructive. In the present study, the authors measured the sugar content of pears using visible and near infrared diffuse reflection spectroscopy. The pretreatment methods of multiplicative scatter correction (MSC), baseline correction, standard normal variate (SNV) transformation, and moving average algorithms were used on the original absorbance spectrum. Results indicate that the absorbance spectra after pretreatment are better than the original absorbance spectra for prediction. Partial least squares (PLS) regression was also used on the original absorbance spectrum and the absorbance spectrum after moving average and baseline correction. It follows that the forecast accuracy of the absorbance spectra after moving average is higher than that of the original absorbance spectra. The models gave good predictions of the sugar content of pears, with corresponding r values of 0.990 8, and standard errors of predictions of 0.019 0.