This review covers developments in non-invasive techniques for quality analysis and inspection of specialty crops, mainly fresh fruits and vegetables, over the past decade up to the year 2010. Presented and discussed in this review are advanced sensing technologies including computer vision, spectroscopy, X-rays, magnetic resonance, mechanical contact, chemical sensing, wireless sensor networks and radio-frequency identification sensors. The current status of different sensing systems is described in the context of commercial application. The review also discusses future research needs and potentials of these sensing technologies. Emphases are placed on those technologies that have been proven effective or have shown great potential for agro-food applications. Despite significant progress in the development of non-invasive techniques for quality assessment of fruits and vegetables, the pace for adoption of these technologies by the specialty crop industry has been slow.

Recent commercialisation of seedless watermelon varieties relies on the guarantee of a high quality product. Several internal defects may deteriorate greatly this fruit: (a) creases and/or large voids in the flesh, (b) overripeness and (c) bruises due to impact. The objective of this research was to develop a feasible non-destructive procedure for detecting these defects in individual fruits, based on acoustic impulse response.\ud A device consisting of a microphone, structural elements and a mechanical impact generator was designed and tested. Good and defective seedless watermelons were tested with the acoustic device. Spectral parameters were examined as potential non-destructive predictors of internal disorders. Waveband magnitude parameters, obtained by summing the magnitude of the spectrum between two frequencies in a specified band width (always including between 40 and 500 Hz), were the acoustic parameters showing the best ability to detect internal disorders

In recent years, Independent Components Analysis (ICA) has proven itself to be a powerful signal-processing technique for solving the Blind-Source Separation (BSS) problems in different scientific domains. In the present work, an application of ICA for processing NIR hyperspectral images to detect traces of peanut in wheat flour is presented. Processing was performed without a priori knowledge of\ud the chemical composition of the two food materials. The aim was to extract the source signals of the different\ud chemical components from the initial data set and to use them in order to determine the distribution of peanut traces in the hyperspectral images. To determine the optimal number of independent component to be extracted, the Random ICA by blocks method was used. This method is based on the repeated calculation of several models using an increasing number of independent components after randomly segmenting the matrix data into two blocks and then calculating the correlations between the signals\ud extracted from the two blocks. The extracted ICA signals were interpreted and their ability to classify\ud peanut and wheat flour was studied. Finally, all the extracted ICs were used to construct a single synthetic\ud signal that could be used directly with the hyperspectral images to enhance the contrast between the peanut and the wheat flours in a real multi-use industrial environment. Furthermore, feature extraction methods (connected components labelling algorithm followed by flood fill method to extract object contours) were applied in order to target the spatial location of the presence of peanut traces. A good visualization of the distributions of peanut traces was thus obtaine

The main objective of this study was to develop a vision system that is able to classify fresh-cut apple slices according to the development of enzymatic browning. The experiment was carried out on ‘Granny Smith’ apple slices stored at 7.5 °C for 9 days (n = 120). Twenty-four samples were analyzed per day: at zero time and after storage for 1, 3, 7 and 9 days, which corresponds to treatments t0, t1, t3, t7 and t9 respectively. Multispectral images were acquired from the samples by employing a 3-CCD camera centered at the infrared (IR, 800 nm), red (R, 680 nm) and blue (B, 450 nm) wavelengths. Apple slices were evaluated visually according to a visual color scale of 1–5 (where 1 corresponds to fresh samples without any browning and 5 to samples with severe discoloration), to obtain a sensory evaluation index (ISE) for each sample. Finally, for each sample and for each treatment, visible (VIS) relative reflectance spectra (360–740 nm) were obtained. In order to identify the most related wavelengths to enzymatic browning evolution, unsupervised pattern recognition analysis of VIS reflectance spectra was performed by principal components analysis (PCA) on the autoscaled data. Maximum loading values corresponding to the B and R areas were observed. Therefore, a classification procedure was applied to the relative histograms of the following monochromatic images (virtual images), which were computed pixel by pixel: (R − B)/(R + B), R − B and B/R. In all cases, a non-supervised classification procedure was able to generate three image-based browning reference classes (BRC): Cluster A (corresponding to the t0 samples), Cluster B (t1 and t3 samples) and Cluster C (t7 and t9 samples). An internal and an external validation (n = 120) were carried out, and the best classifications were obtained with the (R − B)/(R + B) and B/R image histograms (internal validation: 99.2% of samples correctly classified for both virtual images; external validation: 84% with (R − B)/(R + B) and 81% with B/R). The camera classification was evaluated according to the colorimetric measurements, which were usually utilized to evaluate enzymatic browning development (CIE L*a*b* color parameters and browning index, BI) and according to ISE. For both validation phases a*, b*, BI and ISE increased while L* values decreased with image-based class number, thereby reflecting their browning state

The present research is focused on the application of hyperspectral images for the supervision of quality deterioration in ready to use leafy spinach during storage (Spinacia oleracea). Two sets of samples of packed leafy spinach were considered: a) first set of samples was stored at 20°C (E-20) in order to accelerate the degradation process; these samples were measured the day of reception in the laboratory and after two days of storage; b) second set of samples was kept at 10°C (E-10); the measurements were taken throughout the period of storage, beginning the day of reception and repeating the acquisition of images three, six and nine days after. Twenty leaves per testing date of test were analyzed. Hyperspectral images were acquired with a push-broom CCD camera equipped with a spectrograph VNIR (400 to 1000 nm). Calibration set of spectra was extracted from E-20 samples, containing three classes of degradation: class A (optimal quality), class B and class C (maximum deterioration). Reference average spectra were defined for each class. Three models, computed on the calibration set, with a decreasing degree of complexity were compared, according to their ability for segregating spinaches at different quality stages (fresh, with incipient and non visible symptoms of degradation, and degraded): spectral angle mapper distance (SAM), partial least squares discriminant analysis models (PLS-DA), and a non linear index (Leafy Vegetable Evolution, LEVE) combining five wavelengths included among the previously selected by CovSel procedure. In sets E-10 and E-20, artificial images of the membership degree according to the distance of each pixel to the reference classes, were computed assigning each pixel to the closest reference class. The three methods were able to show the evolution degradation of the leaves along the time