This PDF file contains the front matter associated with SPIE Proceedings Volume 6761, including the Title Page, Copyright information, Table of Contents, Introduction (if any), and the Conference Committee listing.

Tests and calibration of sprayers have been considered a very important task for chemicals use reduction in agriculture and for improvement of plant phytosanitary protection. A reliable, affordable and easy-to-use method to observe the distribution in the field is required and the infrared thermoimage analysis can be considered as a potential method based on non-contact imaging technologies. The basic idea is that the application of colder water (10 °C less) than the leaves surface makes it possible to distinguish and measure the targeted areas by means of a infrared thermoimage analysis based on significant and time persistent thermal differences. Trials were carried out on a hedge of Prunus laurocerasus, 2.1 m height with an homogenous canopy. A trailed orchard sprayer was employed with different spraying configurations. A FLIR^TM (S40) thermocamera was used to acquire (@ 50 Hz) thermal videos, in a fixed position, at frame rate of 10 images/s, for nearly 3 min. Distribution quality was compared to the temperature differences obtained from the thermal images between pre-treatment and post-treatment (ΔT)., according two analysis: time-trend of ΔT average values for different hedge heights and imaging ΔT distribution and area coverage by segmentation in k means clustering after 30 s of spraying. The chosen spraying configuration presented a quite good distribution for the entire hedge height with the exclusion of the lower (0-1 m from the ground) and the upper part (>1.9 m). Through the image segmentation performed of ΔT image by k-means clustering, it was possible to have a more detailed and visual appreciation of the distribution quality among the entire hedge. The thermoimage analysis revealed interesting potentiality to evaluate quality distribution from orchards sprayers.

Mining activities, expecially those operated in open air (open pit), present a deep impact on the sourrondings. Such an impact, and the related problems, are directly related to the correct operation of the activities, and usually strongly interact with the environment. Impact can be mainly related to the following issues: high volumes of handled material, ii) generation of dust, noise and vibrations, water pollution, visual impact and, finally, mining area recovery at the end of exploitation activities. All these aspects can be considered very important, and must be properly evaluated and monitored. Environmental impact control is usually carried out during and after the end of the mining activities, adopting methods related to the detection, collection, analysis of specific environmental indicators and with their further comparison with reference thresholding values stated by official regulations. Aim of the study was to investigate, and critically evaluate, the problems related to development of an integrated set of procedures based on the collection and the analysis of remote sensed data in order to evaluate the effect of rehabilitation of land contaminated by extractive industry activities. Starting from the results of these analyses, a monitoring and registration of the environmental impact of such operations was performed by the application and the integration of modern information technologies, as the previous mentioned Earth Observation (EO), with Geographic Information Systems (GIS). The study was developed with reference to different dismissed mine sites in India, Thailand and China. The results of the study have been utilized as input for the construction of a knowledge based decision support system finalized to help in the identification of the appropriate rehabilitation technologies for all those dismissed area previously interested by extractive industry activities. The work was financially supported within the framework of the Project ASIA IT&C - CN/ASIA IT&C/006 (89870) Extract-It "Application of Information Technologies for the Sustainable Management of Extractive Industry Activities" of the European Union.

Rapid and routine identification of foodborne bacteria are considerably important, because of bio- / agro- terrorism threats, public health concerns, and economic loss. Conventional, PCR, and immunoassay methods for the detection of bacteria are generally time-consuming, chemical reagent necessary and multi-step procedures. Fast microbial detection requires minimal sample preparation, permits the routine analysis of large numbers of samples with negligible reagent costs, and is easy to operate. Therefore, we have developed silver colloidal nanoparticle based surface-enhanced Raman scattering (SERS) spectroscopy as a potential tool for the rapid and routine detection of E. coli and L. monocytogenes. This study presents the further results of our examination on S. typhimonium, one of the most commonly outbreak bacteria, for the characteristic bands and subsequent identification.

This paper is concerned with the development of a hyperspectral imaging technique for detecting and identifying one of the most common foodborne pathogens, Campylobacter. Direct plating using agars is an effective tool for laboratory tests and analyses of microorganisms. The morphology (size, growth pattern, color, etc.) of colonies grown on agar plates has been widely used to tentatively differentiate organisms. However, it is sometimes difficult to differentiate target organisms like Campylobacters from other contaminants grown together on the same agar plates. A hyperspectral imaging system operating at the visible and near infrared (VNIR) spectral region from 400 nm to 900 nm was set up to measure spectral signatures of 17 different Campylobacter and non-Campylobacter subspecies. Protocols for culturing, imaging samples and for calibrating measured data were developed. The VNIR spectral library of all 17 organisms commonly encountered in poultry was established from calibrated hyperspectral images. A classification algorithm was developed to locate and identify Campylobacters, non-Campylobacter contaminants, and background agars with 99.29% accuracy. This research has a potential to be expanded to detect other pathogens grown on agar media.

A sandwiched immunoassay consisting of toxin capture by immunomagnetic beads (IMB) and toxin detection by horseradish peroxidase (HRP) linked chemiluminescence was used to follow the production of Shiga-like toxins (SLT) by E. coli O157:H7. The intensity of luminescence generated by the oxidation of luminol-liked compounds was used to represent the concentration of toxins produced. The time-course of SLT production by E. coli O157:H7 under different conditions was investigated. In pure culture, optimal generation of SLT showed a significant delay than the steady state of cell growth. In mixed cultures of SLT producing E. coli O157:H7 and non-SLT producing E. coli K-12 strain, the production of toxins was substantially decreased. However, the growth of E. coli O157:H7 was not affected by the presence of K-12 strain. This decrease in SLT production was also observed in radiation-sterile ground beef. In regular ground beef that might contain numerous other bacteria, the growth of E. coli O157:H7 in EC media was not significantly affected but the lowered production of SLT was observed. The results showed that mechanism of inducing SLT production was complex with both the growth time and growth environment could influence SLT production. The addition of homo-serine lactone to the growth media enhanced the production of SLT. Thus, possibly cell-cell communication may have a role in SLT production by E. coli O157:H7.

Wheat durum pasta (spaghetti in particular) can be considered as the most typical Italy's food product. Many small or craft pasta factories realize different quality product regarding the use of biological wheat and the application of mild (lower drying temperature) or traditional (bronze draw-plate) technologies, in competition with large industrial enterprises. The application of higher quality standards increases the producing cost and determines higher pasta prices. In order to setup a reliable easy-to-use methodology to distinguish different production technology approaches, spectrophotometric visible and near-infrared (VIS-Nir) techniques were applied on the intact pasta. Eighteen samples of commercial brand spaghetti classified in five different quality production factors (Full industrial - Teflon-drawn, high temperature-short time drying -; semolina from organic cultivations; bronze-drawn treatment; low temperature - long time drying; traditional high quality pasta - bronze-drawn and low temperature drying treatments-) were analyzed by three different spectrometric techniques: a VIS (400 - 700 nm) spectral imaging, a Nir (1000-1700 nm) spectral imaging - both of them acquiring reflected spectral images of spaghetti bundle - and a portable VIS-Nir system (400-800 nm), working with an interactance probe on single spaghetti string. Principal component analysis (PCA) and partial least square regressions (PLS) were performed on about 1500 spectral arrays, to test the ability of the systems to distinguish the different pasta products (commercial brands). Reflectance visible data presented highest percentage of correct classification: 98.6% total value, 100% for high quality spaghetti (bronze-drawn and/or low temperature drying). NIR reflectance and VIS-NIR interactance systems presented 85% and 70% of entire correct classification while for high quality pasta the percentages rise up to 75% and 83%

Paddy rice is a staple food in China and it's growth monitoring, acreage extraction and yield estimate are of far reaching importance. It is difficult to apply conventional remote sensing technique for obtaining precise information on paddy planting and growth, for rice bowls are mostly distributed over rainy regions in China. The radar image is unlimited by cloud, rain and fog, and could proceed all weather operation and obtain more stable data, therefore it could be used for paddy monitoring. Making use of Envisat's ASAR data and NOAA data in 2004, paddy's backward-scattering characteristics with different polarizations were studied in this paper. To combine multi-temporal radar data with one view ETM image, paddyfield of experimental area in Hongze of Jiangsu Province was classified. Results show that 1) characteristics of paddy's hh and vv polarizations vary from stage to stage and vv polarization is more sensitive. The polarization ratio hh / vv of paddy during metaphase is apparently higher than other objects'. 2) paddy's polarization ratio hh / vv and growth vigor closely relate to each other , thereof two empirical time-domain models of backward- scattering were established, wherewith to estimate number of days after transplanting and growing season. 3) hh and ratio hh / vv are both well correlated with NDVI. 4) hh polarization data could be used for information extraction of towns and water bodies, and the hh / vv image in metaphase for partition of paddy from other objects. The recognition accuracy being ninety percent over, multi-temporal and -polarization radarsat data are of predominance and potential for paddy growth and/or acreage monitoring.

One of the most common molds that infects the seeds of small cereals worldwide, such as wheat, is Fusarium Head Blight (FHB). The mycotoxin, deoxynivalenol (also known as DON or vomitoxin) is often produced by this mold, which, upon ingestion, causes health problems to not only livestock (especially non-ruminants), but to humans as well. In the United States, the FDA has established advisory levels for DON in food and feeds, a practice that is likewise conducted by most countries of the world. Our previous research has shown that commercial high-speed optical sorters are on average 50 percent efficient at the removal of mold-damaged kernels; however, under more careful control in the laboratory, this efficiency can rise to 95 percent or better. Ongoing research is examining the potential to achieve the higher efficiencies at conditions that are more akin to those of commercial processing. For example, multispectral information is collected on single kernels in freefall at the sub-millisecond level. Knowledge gained from this research will provide design criteria for improvement of high-speed optical sorters for reduction of DON in raw cereals commodities, as well as in finished food products.

Sarcomere is the fundamental functional unit in skeletal muscle for force generation. In addition, sarcomere structure is also an important factor that affects the eating quality of muscle food, the meat. The sarcomere structure is altered significantly during rigor mortis, which is the critical stage involved in transforming muscle to meat. In this paper, we investigated optical scattering changes during the rigor process in Sternomandibularis muscles. The measured optical scattering parameters were analyzed along with the simultaneously measured passive tension, pH value, and histology analysis. We found that the temporal changes of optical scattering, passive tension, pH value and fiber microstructures were closely correlated during the rigor process. These results suggested that sarcomere structure changes during rigor mortis can be monitored and characterized by optical scattering, which may find practical applications in predicting meat quality.

A clear understanding of how light propagation in muscle is important for developing optical methods for muscle characterization. We investigated photon migration in muscle by imaging the optical reflectance from fresh prerigor skeletal muscles. We found the acquired reflectance patterns can not be described using existing theories. In order to quantify the equi-intensity contours of acquired reflectance images, we developed a numerical fitting function. Using this model, we studied the changes of reflectance profile during stretching and rigor process. The observed unique anisotropic features diminished after rigor completion. These results suggested that muscle sarcomere structures played important roles in modulating light propagation in whole muscle. To explain the observed patterns, we incorporated the sarcomere diffraction in a Monte Carlo model and we showed that the resulting reflectance profiles quantitatively resembled the experimental observation.

Phototropic microalgae have a large potential for producing valuable substances for the feed, food, cosmetics, pigment, bioremediation, and pharmacy industries as well as for biotechnological processes. Today it is estimated that the microalgal aquaculture worldwide production is 5000 tons of dry matter per year (not taking into account processed products) making it an approximately $1.25 billion U.S. per year industry. For effective observation of the photosynthetic growth processes, fast on-line sensor systems that analyze the relevant biological and technical process parameters are preferred. The optical properties of the microalgae culture influence the transport of light in the photobioreactor and can be used to extract relevant information for efficient cultivation practices. Microalgae cultivation media show a combination of light absorption and scattering, which are influenced by the concentrations and the physical and chemical properties of the different absorbing and scattering species (e.g. pigments, cell components, etc.). Investigations with frequency domain photon density waves (PDW) allow for the examination of absorption and scattering properties of turbid media, namely the absorption and reduced scattering coefficient. The reduced scattering coefficient can be used to characterize physical and morphological properties of the medium, including the cell concentration, whereas the absorption coefficient correlates with the pigment content. Nannochloropsis oculata, a single-cell species of microalgae, were examined in a nutrient solution with photon density waves. The absorption and reduced scattering coefficients were experimentally determined throughout the cultivation process, and applied to gain information about the cell concentration and average cell radius.

Oxytetracycline (OTC) residue in honey is detected using a portable analyzer designed to specifically target tetracycline (TC) drugs based on europium-sensitized luminescence (ESL). A 385 nm light emitting diode (LED) is used as the excitation source and a photomultiplier tube as the light detector. OTC is extracted from honey and cleaned up by solid phase extraction (SPE) using Strata X-WC weak cation exchange cartridges. To the eluate Eu(III) is added to form a Eu-TC chelate at pH 8.5. Efficient intrachelate energy transfer allows sensitive OTC detection at λ_ex=385 nm and λ_em=610 nm. After a 25-µs time delay, the ESL signal is integrated over a 25-1000 µs interval. The signal intensity reveals a linear relationship (R²=0.972) to OTC concentrations in the 10-200 ng/g range. The limit-of-detection is 6.7 ng/g with an average 5.8% relative standard deviation. The background signal corresponds to ~10 ppb. This instrumentation and method combination enables field analysis that is especially useful for beekeeping industry.

Monitoring of nutrient status of crops is essential for better management of crop production. Nitrogen is one of the most important elements in fertilizer for the growth and yield of vegetable crops. In this study, nitrogen content of cabbage seedlings was evaluated using hyper-spectral images. Cabbage seedlings, cultured at five nitrogen fertilization levels, were planted in the 128-cell plug trays and grown in a phytotron at National Taiwan University. The images, ranged from 410 to 1090 nm, of cabbage seedlings were analyzed by a hyper-spectral imaging system consisting of CCD cameras with liquid crystal tunable filters (LCTF), which was developed in this study. The digital images of seedling canopies were processed including image segmentation, gray level calibration and absorbance conversion. Models including modified partial least square regression (MPLSR), step-wise multi-linear regression (SMLR) and artificial neural network with cross-learning strategy (ANN-CL) were developed for the determination of the nitrogen content in cabbage seedlings. The three significant wavelengths derived from SMLR model are 470, 710, and 1080; and the best result is obtained by ANN-CL model, in which r_c=0.89, SEC=6.41 mg/g, r_v=0.87, and SEV=6.96 mg/g. The ANN-CL model is more suitable for the remote sensing in precision agriculture applications because not only its model accuracy but also only 3 wavelengths are needed.

The objective of this research was to measure the absorption and reduced scattering coefficients (μ_a and μ_s', respectively) of apples via a spatially-resolved hyperspectral imaging technique and relate them to fruit firmness and soluble solids content (SSC). Spatially-resolved hyperspectral images were acquired from 600 'Golden Delicious' apples, and values for μ_a and μ_s' were determined using an inverse algorithm to fit the diffusion theory model to the spectral scattering profiles over 500-1000 nm. There were two predominant peaks in the absorption spectra around 675 nm and 970 nm due to the presence of chlorophyll and water in the fruit, respectively. Spectra of μ_s' decreased monotonically with the increasing wavelength. Both μ_a and μ_s' were correlated with fruit firmness, with the correlation coefficient (r) of 0.82 and 0.80, respectively. Values of μ_a were also correlated with the SSC with r = 0.70. The combined data of μ_a and μ_s' were able to predict fruit firmness with r = 0.88 and the standard error of prediction (SEP) of 5.66 N, and SSC with r = 0.82 and SEP = 0.75%. This research demonstrated the potential of using spectral absorption and scattering properties to evaluate internal quality attributes of horticultural products.

Citrus grove change detection is of great importance to citrus production inventory monitoring. Using remotely sensed imagery to detect the land use and land coverage is one of the most widely-used, cost-effective approaches. However, there is little published research on citrus grove change detection using remotely sensed multi-spectral imagery, especially for those acquired by heterogeneous sensors. The purpose of this paper is to investigate the effectiveness of the citrus change detection based on the histogram matching normalization to the heterogeneously sensed imagery. In this paper, it is found that different reference image and band selection will result in different normalization performance. Based on this finding, a concept of finding optimal reference image and best spectral band for normalization in terms of the minimum Manhattan distance measure is presented. In this paper, the comparison of change detection results of unnormalized and histogram matching normalized images is presented. The experimental results show that histogram matching normalization significantly improves the image differencing based change detection results of the heterogeneously sensed citrus images, and the optimal reference image and band found with proposed optimization algorithm gives the best change detection results.

Automated imaging systems offer the potential to inspect the quality and safety of fruits and vegetables consumed by the public. Current automated inspection systems allow fruit such as apples to be sorted for quality issues including color and size by looking at a portion of the surface of each fruit. However, to inspect for defects and contamination, the whole surface of each fruit must be imaged. The goal of this project was to develop an effective and economical method for whole surface imaging of apples using mirrors and a single camera. Challenges include mapping the concave stem and calyx regions. To allow the entire surface of an apple to be imaged, apples were suspended or rolled above the mirrors using two parallel music wires. A camera above the apples captured 90 images per sec (640 by 480 pixels). Single or multiple flat or concave mirrors were mounted around the apple in various configurations to maximize surface imaging. Data suggest that the use of two flat mirrors provides inadequate coverage of a fruit but using two parabolic concave mirrors allows the entire surface to be mapped. Parabolic concave mirrors magnify images, which results in greater pixel resolution and reduced distortion. This result suggests that a single camera with two parabolic concave mirrors can be a cost-effective method for whole surface imaging.

The classification of walnuts shell and meat has a potential application in industry walnuts processing. A dark-field illumination method is proposed for the inspection of walnuts. Experiments show that the dark-field illuminated images of walnut shell and meat have distinct text patterns due to the differences in the light transmittance property of each. A number of rotation invariant feature analysis methods are used to characterize and discriminate the unique texture patterns. These methods include local binary pattern operator, wavelet analysis, circular Gabor filters, circularly symmetric gray level co-occurrence matrix and the histogram-related features. A recursive feature elimination method (SVM-RFE), is used to remove uncorrelated and redundant features and to train the SVM classifier at the same time. Experiments show that, by using only the top six ranked features, an average classification accuracy of 99.2% can be achieved.

Contaminant detection on chicken carcasses is an important product inspection application. The four contaminant types of interest contain three types of feces from different gastrointestinal regions (duodenum, ceca, and colon) and ingesta (undigested food) from the gizzard. Use of automated or semi-automated inspection systems for detecting fecal contaminant regions is of great interest. Hyperspectral data provided by ARS (Athens, GA) were used to examine detection of contaminants on carcasses. We address quasi-optimal algorithms for selecting a set of spectral bands (wavelengths) in hyperspectral data for on-line contaminant detection (feature selection). We introduce our new improved forward floating selection (IFFS) algorithm and compare its performance to that of other state-of-the-art feature selection algorithms. Our initial results indicate that our method gives an excellent detection rate and performs better than other feature selection algorithms. We also show that combination feature selection algorithms perform worse.

We consider new methods to select useful sets of ratio features in hyperspectral data to detect contaminant regions on chicken carcasses using data provided by ARS (Athens, GA). A ratio feature is the ratio of the response at each pixel for two different wavebands. Ratio features perform a type of normalization and can thus help reduce false alarms, if a good normalization algorithm is not available. Thus, they are of interest. We present a new algorithm for the general problem of such feature selection in high-dimensional data. The four contaminant types of interest are three types of feces from different gastrointestinal regions (duodenum, ceca, and colon) and ingesta (undigested food) from the gizzard. To select the best two sets of ratio features from this 492-band HS data requires an exhaustive search of more than seven billion combinations of two sets of ratio features, which is very excessive. Thus, we propose our new fast ratio feature selection algorithm that requires evaluation of a much fewer number of sets of ratio features and is capable of giving quasi-optimal or optimal sets of ratio features. This new feature selection method has not been previously presented. It is shown to offer promise for an excellent detection rate and a low false alarm rate for this application. Our tests use data with different feed types and different contaminant types.

Multispectral scattering is effective for nondestructive prediction of fruit firmness. However, the established prediction models for multispectral scattering are variety specific and may not perform appropriately for fruit harvested from different orchards or at different times. In this research, a recursive least squares method was proposed to update the existing prediction model by adding samples from a new population to assure good performance of the model for predicting fruit from the new population. Multispectral scattering images acquired by a multispectral imaging system from Golden Delicious apples that were harvested at the same time but had different postharvest storage time periods were used to develop the updating method. Radial scattering profiles were described by the modified Lorentzian distribution (MLD) function with four profile parameters for eight wavelengths. Multi-linear regression was performed on MLD parameters to establish prediction models for fruit firmness for each group. The prediction model established in the first group was then updated by using selected samples from the second group, and four different sampling methods were compared and validated with the rest apples. The prediction model corrected by the model-updating method gave good firmness predictions with the correlation coefficient (r) of 0.86 and the standard error of prediction (SEP) of 6.11 N. This model updating method is promising for implementing the spectral scattering technique for real-time prediction of apple fruit firmness.

In order to sell grain lots in European countries, seed industries need a government certification. This certification requests purity testing, seed counting in order to quantify specified seed species and other impurities in lots, and germination testing. These analyses are carried out within the framework of international trade according to the methods of the International Seed Testing Association. Presently these different analyses are still achieved manually by skilled operators. Previous works have already shown that seeds can be characterized by around 110 visual features (morphology, colour, texture), and thus have presented several identification algorithms. Until now, most of the works in this domain are computer based. The approach presented in this article is based on the design of dedicated electronic vision machine aimed to identify and sort seeds. This machine is composed of a FPGA (Field Programmable Gate Array), a DSP (Digital Signal Processor) and a PC bearing the GUI (Human Machine Interface) of the system. Its operation relies on the stroboscopic image acquisition of a seed falling in front of a camera. A first machine was designed according to this approach, in order to simulate all the vision chain (image acquisition, feature extraction, identification) under the Matlab environment. In order to perform this task into dedicated hardware, all these algorithms were developed without the use of the Matlab toolbox. The objective of this article is to present a design methodology for a special purpose identification algorithm based on distance between groups into dedicated hardware machine for seed counting.

In the majority of food and feed, due to the microscopic spatial changes in the refractive index, visible (VIS) and near infrared (NIR) light undergoes multiple scattering events and the overall light distribution is determined more by scattering rather than absorption. Conventional steady state VIS/NIR reflectance spectroscopy can provide information on light attenuation, which depends both on light absorption and light scattering, but cannot discriminate these two effects. On the contrary, time-resolved reflectance spectroscopy (TRS) provides a complete optical characterisation of diffusive media in terms of their absorption coefficient and reduced scattering coefficient. From the assessment of the absorption and reduced scattering coefficients, information can then be derived on the composition and internal structure of the medium. Main advantages of the technique are the absolute non-invasiveness, the potentiality for non-contact measurements, and the capacity to probe internal properties with no influence from the skin. In this work we review the physical and technical issues related to the use of TRS for nondestructive quality assessment of fruit and vegetable. A laboratory system for broadband TRS, based on tunable mode-locked lasers and fast microchannel plate photomultiplier, and a portable setup for TRS measurements, based on pulsed diode lasers and compact metal-channel photomultiplier, will be described. Results on broadband optical characterisation of fruits and applications of TRS to the detection of internal defects in pears and to maturity assessment in nectarines will be presented.

In this paper, 104 samples of Chinese rice wines of the same variety (Shaoxing rice wine), collected in three winery ("guyuelongshan", "pagoda" brand, "kuaijishan"), three brewed years (2002, 2004, 2004-2006) were analyzed by near-infrared transmission spectroscopy between 800 and 2500 nm. The spectral differences were studied by principal components analysis (PCA), and Classifications, according the brand, were carried out by discriminant analysis (DA) and partial least squares discriminant analysis (PLSDA). The DA model gained a total accuracy of 94.23% and when used to predict the brand of the validation set samples, a better result, correctly classified all of the three kinds of Chinese rice wine up to 100%, are obtained by PLSDA model. The work reported here is a feasibility study and requires further development with considerable samples of more different brands. Further studies are needed in order to improve the accuracy and robustness, and to extend the discrimination to other Chinese rice wine varieties or brands.

The potential of near-infrared (NIR) transmittance spectroscopy to nondestructively detect soluble solids contents (SSC) and pH in tomato juices was investigated. A total of 200 tomato juice samples were used for NIR spectroscopy analysis at 800-2400 nm using FT-NIR spectrometer. Multiplicative signal correcton (MSC), the first and second derivative were applied for preprocessing spectral data. The relationship between SSC, pH and FT-NIR spectra of tomato juice was analyzed via partial least-squares (PLS) regression, respectively. PLS regression models for SSC and pH in tomato juices show the high accuracy. The correlation coefficient (r), root mean square error of calibration (RMSEC) and root mean square error of validation (RMSEP), root mean square error of cross-validation (RMSECV) for SSC were 0.91582, 0.0703, 0.150 and 0.138, respectively, whereas those values for pH were 0.8997, 0.0333, 0.0316 and 0.0489, respectively. It is concluded that the NIR transmittance spectroscopy is promising for the fast and nondestructive detection of chemical components in tomato juices.

White peach is a famous peach variety for its super-quality and high economic benefit. It is originally planted in Yuandong Villiage, Jinhua County, Zhejiang province. By now, it has been planted in many other places in southeast of China. However, peaches from different planting areas have dissimilar quality and taste, which result in different selling price. The objective of this research was to discriminate peaches from different planting areas by using near-infrared (NIR) spectra and chemometrics methods. Diffuse reflectance spectra were collected by a fiber spectrometer in the range of 800-2500 nm. Discriminant analysis (DA), soft independent modeling of class analogy (SIMCA), and discriminant partial least square regression (DPLS) methods were employed to classify the peaches from three planting areas 'Jinhua', 'Wuyi', and 'Yongkang' of Zhejiang province. 360 samples were used in this study, 120 samples per planting area. The classifying correctness were above 92% for both DA and SIMCA mdoels. And the result of DPLS model was slightly better. By using DPLS method, two 'Jinhua' peaches, three 'Wuyi' peaches, and three 'Yongkang' peaches were misclassified, the accruacy was above 95%. The results of this study indicate that the three chemometrics methods DA, SIMCA, and DPLS are effective for discriminating peaches from different planting areas based on NIR spectroscopy.

Near-infrared (NIR) transmittance spectroscopy combined with several chemometrical techniques was investigated to study the oxidation process during storage in tomato juices. A total of 100 tomato juice samples were used for NIR spectroscopy analysis at 800-2400 nm using FT-NIR spectrometer. The spectrum of each tomato juice was collected twice: the first time as soon as the tomatoes were squeezed, centrifuged, filtered and the tomato juice had not undergone any oxidation process and the second measurement was taken after a month. Principal component analysis (PCA) and partial least squares discriminant analysis (PLSDA) were applied to discriminate between the two groups of spectra. The results show that differences between tomato juices before and after the storage period do exist attributed to changes in certain compounds of juice and excellent classification can be obtained after optimizing spectral pretreatment.

This paper is aimed at investigating the possibility of sorting rice seeds by rapid techniques. Machine vision and dielectric separation were involved to determine external and internal quality of rice seeds. A conceptual rapid seed sorter is proposed. Two varieties of rice seeds planted and harvested in different years were involved in the experiments. Using morphological and color features gave a highly acceptable classification of normal and defective seeds. Dielectric parameters can be used to classify rice seeds into high vigor and low vigor. Combination of appearance characteristics and dielectric properties provide comprehensive response of seed quality. A highly acceptable defects classification and vigor improvement were achieved when the principle prototype was implemented for all the samples to test the adaptability. The good adaptability of machine vision and dielectric separation indicate the potential to determine quality of rice seeds rapidly. This paper presents the significant elements of the conceptual prototype and emphasizes the important aspects of the image processing and dielectric separation techniques.

Applications of machine vision in automated inspection and sorting of fruits have been widely studied by scientists and. Preprocess of the fruit image is needed when it contain much noise. There are many methods for image denoise in literatures and can acquire some nice results, but which will be selected from these methods is a trouble problem. In this research, total variation (TV) and shock filter with diffusion function were introduced, and together with other 6 common used denoise method s for different type noise type were tested. The result demonstrated that when the noise type was Gaussian or random, and SNR of original image was over 8,TV method can achieve the best resume result, when the SNR of original image was under 8, Winner filter can get the best resume result; when the noise type was salt pepper, median filter can achieve the best resume result

High quality standards in modern wine production strictly depend of the choice of optimal maturity stage of grapes for the wine-making. Different chemical parameters of grape juice and peel were usually analysed in order to establish the optimal time of harvest. Aim of the study was to test the capability of a spectrophotometric visible and near-infrared (VIS-Nir) portable non-destructive system, to estimate chemical parameters to establish the optimal harvesting period. Spectral acquisition on wine grapes were made at three times before harvesting (18, 15, 9 days) and at the harvest for three different cultivars: Cabernet Sauvignon, San Giovese, Merlot. Trials were conducted in a vineyard located in South Tuscany, typical production area of Morellino di Scansano wine (Marchesi de' Frescobaldi producer). A VIS-Nir spectrometer - wavelength range 400 - 1000 nm, 3 nm bandwidth - equipped with a reflectance optical fiber probe (4 mm diameter) was used to estimate reductant sugars, total acidity, pH, potential anthocyanins and maturity index (sugar/acidity ratio) in whole wine grapes. A partial least square regression was performed for the different sampling times, including more than 3000 spectral measurements. Estimation of chemical parameters were performed with different standard error of prevision (SEP) and correlation coefficient (R): near to SEP=10% in respect to the average of the observed value and R=70%. Results showed that VIS-NIR reflectance was a suitable non-destructive method for monitoring the wine grapes maturity stage.

Posidonia oceanica (L.) is an endemic phanerogam of the Mediterranean Sea. It lives between 0.2 and 40 m depth and make up extensive meadows that play a fundamental role in the marine coast ecosystem. Near the coasts at higher anthropic pressure, Posidonia meadows present both quality and quantity damages (regression) due to the mechanical operations on the seabed (anchoring, drag netting, pipe lines) and the sea pollution. Nowadays, the seagrass regression is monitored by different systems: aereophotografic, side scan sonar, underwater television camera, direct underwater visual inspection. Scientific community is looking for to develop monitoring systems more reliable, rapid and non invasive. Aim of this study is to evaluate the application of a new spectrophotometric imaging system based on the acquisition of reflectance spectral images with a good optical (250 Kpixels) and spectral resolution (spectral range 400-970 nm, a total of 115 single wavelength, 5 nm step each one). First trials were made on Posidonia's leafs to evaluate the system capacity to recognize spectral differences between samples picked up at two different depths (0.3 - 4 m). High discrimination percentage (90%) were found between leaf samples as function of the different depths, analyzing the spectral data by Partial Least Squares model. Forward activities will stress the system capability also to evaluate different phenol concentrations on Posidonia leaves, an important index of physiologic vegetal damage, through direct underwater spectrophotometric monitoring.

Rabbit meat is for its nutritional characteristics a food corresponding to new models of consumption. Quality improvement is possible integrating an extensive organic breeding with suitable rabbit genetic typologies. Aim of this work (financed by a Project of the Lazio Region, Italy) was the characterization of rabbit meat by a statistic model, able to distinguish rabbit meat obtained by organic breeding from that achieved industrially. This was pursued through the analysis of spectral data and colorimetric values. Two genetic typologies of rabbit, Leprino Viterbese and a commercial hybrid, were studied. The Leprino Viterbese has been breeded with two different systems, organic and industrial. The commercial hybrid has been bred only industrially because of its characteristics of high sensibility to diseases. The device used for opto-electronic analysis is a VIS-NIR image spectrometer (range: 400-970 nm). The instrument has a stabilized light, it works in accordance to standard CIE L*a*b* technique and it measures the spectral reflectance and the colorimetric coordinates values. The statistic data analysis has been performed by Partial Least Square technique (PLS). A part of measured data was used to create the statistic model and the remaining data were utilized in phase of test to verify the correct model classification. The results put in evidence a high percentage of correct classification (90%) of the model for the two rabbit meat classes, deriving from organic and industrial breeding. Moreover, concerning the different genetic typologies, the percentage of correct classification was 90%.

The work focused the application of an image analysis technique to determine corn leaves morphology as objective indicator of the growth performance of corn (Zea mays) resulting from the urban residual fertilization. The analyses were related to six fertilization plots: original soil; chemical fertilizer (160 and 200 kg ha-1 of nitrogen); organic fertilizer (32 t ha-1) and two different doses of urban residues (sewage sludges) (7.5 and 22.5 t ha-1, this last amount corresponds to is the maximum level permitted from the Italian law in three year of fertilization). Those tests were realized by full randomized plots, with two three repetitions for each treatment. Measurements were performed for the first year of the trials in the period proximate to harvest (Rome, Italy - July 2000). Four plants for each plot were harvested and stripped of all leaves, whose RGB images were acquired by a digital photo camera (Kodak Ltd). Image analysis was performed first through the separation of RGB channels into single monochromatic 8-bit distribution, than the blue channel images, the most informative, were then submitted to enhancement, low pass filtering to reduce noise, threshold of binarization (based on statistical parameter affected on Gaussian grey levels distribution), binary morphology and object measurement. For ach single leaf the length, the width, the area were measured. The test results indicated positive and significant responses in relation between the crop growth (leaves area, length and width greater) and the different doses of urban residues (sewage sludges).