Search

Artificial vision has wide-ranging applications in the food sector; it is easy to use, relatively low cost and allows to conduct rapid non-destructive analyses. The aim of this study was to use artificial vision techniques to control and model the coffee roasting process. Samples of Castillo variety coffee were used to construct the roasting curve, with captured images at different times. Physico-chemical determinations, such as colour, titratable acidity, pH, humidity and chlorogenic acids, and caffeine content, were investigated on the coffee beans. Data were processed by (i) Principal component analysis (PCA) to observe the aggrupation depending on the roasting time, and (ii) partial least squares (PLS) regression to correlate the values of the analytical determinations with the image information. The results allowed to construct robust regression models, where the colour coordinates (L*, a*), pH and titratable acidity presented excellent values in prediction (R2 Pred 0.95, 0.91, 0.94 and 0.92). The proposed algorithms were capable to correlate the chemical composition of the beans at each roasting time with changes in the images, showing promising results in the modelling of the coffee roasting process. [ABSTRACT FROM AUTHOR]

Virtualization plays a critical role in enriching the user experience in Virtual Reality (VR) by offering heightened realism, increased immersion, safer navigation, and newly achievable levels of interaction and personalization, specifically in indoor environments. Traditionally, the creation of virtual content has fallen under one of two broad categories: manual methods crafted by graphic designers, which are labor-intensive and sometimes lack precision; traditional Computer Vision (CV) and Deep Learning (DL) frameworks that frequently result in semi-automatic and complex solutions, lacking a unified framework for both 3D reconstruction and scene understanding, often missing a fully interactive representation of the objects and neglecting their appearance. To address these diverse challenges and limitations, we introduce the Virtual Experience Toolkit (VET), an automated and user-friendly framework that utilizes DL and advanced CV techniques to efficiently and accurately virtualize real-world indoor scenarios. The key features of VET are the use of ScanNotate, a retrieval and alignment tool that enhances the precision and efficiency of its precursor, supported by upgrades such as a preprocessing step to make it fully automatic and a preselection of a reduced list of CAD to speed up the process, and the implementation in a user-friendly and fully automatic Unity3D application that guides the users through the whole pipeline and concludes in a fully interactive and customizable 3D scene. The efficacy of VET is demonstrated using a diversified dataset of virtualized 3D indoor scenarios, supplementing the ScanNet dataset. [ABSTRACT FROM AUTHOR]

A technique that combines the spatial resolution of a 3D structured-light (SL) imaging system with the spectral analysis of a hyperspectral short-wave near infrared system was developed for freshness predictions of gilthead sea bream on the first storage days (Days 0-6). This novel approach allows the hyperspectral analysis of very specific fish areas, which provides more information for freshness estimations. The SL system obtains a 3D reconstruction of fish, and an automatic method locates gilthead's pupils and irises. Once these regions are positioned, the hyperspectral camera acquires spectral information and a multivariate statistical study is done. The best region is the pupil with an R² of 0.92 and an RMSE of 0.651 for predictions. We conclude that the combination of 3D technology with the hyperspectral analysis offers plenty of potential and is a very promising technique to non destructively predict gilthead freshness. [ABSTRACT FROM AUTHOR]

The capability of the VIS/SW-NIR (visible/short wave near infrared) hyperspectral imaging system to characterize the heat treatment process of cake wheat flour was studied. Combinations of heat treatments of flour were run at different temperatures (80, 100 and 130 °C) and for various times (10, 20 and 30 min). The resulting treated flours were analyzed by the imaging technique. The hyperspectral results, studied by multivariate statistical methods, showed a pattern evolution of the flours treated by different heat treatments. The wavelengths that contributed the most, and implied in the differentiations, were detected. The selection of wavelengths allowed us to optimize the analysis, which reduced from 54 to 6 wavelengths. To ensure that the VIS/SW-NIR information depended on the heat treatment influence on flours, cakes were produced and characterized according to height, mass loss during the baking process, crumb structure and textural properties. The VIS/SW-NIR imaging analysis was capable of following the changes that occurred during the different heat treatments of flours. VIS/SW-NIR was applied to determine and adjust the heat treatment process variables to improve the features of flours during the cake production process. [ABSTRACT FROM AUTHOR]

Ready-to-eat foods that does not receive a heat treatment before being consumed can be at risk of foodborne hazards and spoilage, so it would be of great interest to have a method for monitoring their safety. This work expands on and enhances previous successfully studies with hyperspectral imaging in the SW-NIR range. Specifically, a k-nearest-neighbours model was developed to classify the salmon tissue into white myocommata stripes (fat) and muscle (lean) tissue. Partial Least Squares models developed confirm that a spatial segmentation should be performed before a shelf life model can be calculated. Employing the fat spectra and only the 7 most correlated wavelengths, a support vector machine model was calculated to classify into days 0, 10, 20, 40 and 60 with 87.2% prediction accuracy. These results make the method developed very promising as a non-destructive method to analyse the shelf life of vacuum-packed chilled smoked salmon fillets. [ABSTRACT FROM AUTHOR]

Characterisation of wheat flour destined for the bread-making process was studied by using hyperspectral image analyses based on Short-wave NIR imaging (SW-NIR). The first step was to characterise wheat flour batches by considering physicochemical, rheological and fermentation kinetic parameters. The second step was to capture the hyperspectral images of each batch. The results showed some significant differences in fermentation capacity in relation to gluten and its protein sub fractions. However, no significant rheological differences were observed. The acquired images were firstly analysed by principal component analysis, where a clustering tendency was observed. Then linear correlation maps were done to study the relationship between the SW-NIR spectra response and the physicochemical, rheological and fermentation kinetic parameters. The results showed an important correlation between the SW-NIR spectra (760–900 nm) and the fermentation capacity parameters of the batches, as well as gluten and its sub fraction proteins. The SW-NIR imaging analysis enabled to obtain useful information about wheat flour behaviour related to the fermentation process. Thus the studied technique may represent a useful, rapid non-destructive method to help characterise wheat flours with non-different rheological parameters by considering their behaviour during the fermentation process. [ABSTRACT FROM AUTHOR]

The bread-making process is a set of operations where could be relevant to use monitoring methods. Specifically, the fermentation phase is a crucial step in which the quality of the product can be affected. Several methods have been developed to monitor this stage based on different technologies. The aim of this study was to analyze and obtain information about the internal structure of bread dough during the fermentation process using a 3D vision system based on Structured Light (SL). The differences about fermentation behavior of two wheat flours classified as “high strength flour” by company providing were studied. The parameters of the internal structure of the baked product (final bubble size and their population density) were analyzed with 2D image segmentation. An important correlation ( r = 0.865) was observed between the 3D and 2D information, specifically between the transversal area and height (3D), and final bubble size and number of bubbles (2D). Although at the end of the dough fermentation process ( Tf ) the area ( A ) and maximum height ( H ) were different, the relationship between both parameters were similar, reaching a similar bubble size as a consequence of coalescence phenomena, independent of the bubble growth rate. This could be a base for the development of prediction models and devices to monitor the fermentation phase of the bread-making process. [ABSTRACT FROM AUTHOR]

A non-destructive Vis-NIR spectroscopy (400-1000 nm) method was developed to evaluate the loss of freshness of sliced and commercially packaged cooked ham and turkey ham without any sample manipulation. The spectra were recorded at 0, 30, 40, and 60 days using a camera, spectral filter (400-1000 nm) and a halogen floodlighting system which had been were developed and calibrated for the purpose. Physico-chemical, biochemical, and microbiological properties such as pH, total volatile basic nitrogen (TVB-N), ATP breakdown compounds, and colony-forming units were determined to predict the degradation of freshness. The image spectra obtained from visible and SW-NIR spectroscopy were related to the storage time of the samples. A PLS-DA model was developed independently for packaged or unpackaged samples using the second derivative of the spectra. Mean R²prediction obtained for cooked ham was 0.915 and 0.949 for Turkey ham. The technique developed could be applied to monitoring the freshness of commercial packed cooked ham and turkey ham as a non-destructive technique. Further studies will be needed to check the spectra obtained from samples of different commercial brands in order to evaluate more precisely the efficiency of the method. [ABSTRACT FROM AUTHOR]

Highlights: [•] A 3D vision technique for continuous monitoring of dough fermentation is proposed. [•] Ten samples of wheat flour with the same physicochemical properties were analyzed. [•] Wheat flour samples present different behavior during fermentation process. [•] The information obtained by technique is able to characterize the flour behaviors. [Copyright &y& Elsevier]

Abstract: A comparison between the techniques Time of flight (TOF) and Structured Light (SL), for in-line process, was carried out to determine the volume of several food types with different compositions, structures and dimensions; 2 meat tissue (lean meat and pork fat) and 3 vegetable tissue (potato, apple and avocado). The volume obtained was compared with that determined by physical measurements, employing two statistical methods (a Bland–Altman study and partial least square analysis). Results showed that Structured Light (SL) was a better technique than Time of flight (TOF) for determining volume. The TOF technique was affected by factors which were more influential when the S/V ratio (surface of the sample exposed to the camera and sample volume) increased. SL was slightly affected by the composition of the sample. Fat content and the level of unsaturated fatty acids could be the reason for the reflection of the laser light on the surface of the samples thereby reducing the accuracy. Even so, the values of R 2 and RMSE of cross-validation, for the worst fit, demonstrated the quality of the SL technique. [Copyright &y& Elsevier]

Highlights: [•] Hyperspectral information in the range of SW-NIR used to detect expired salmon. [•] Technique used does not require manipulating the packed salmon samples. [•] Discrimination of fish tissues using quartile segmentation in RGB images. [•] Comparative of our method to filter hyperspectral information with classic filters. [Copyright &y& Elsevier]

The maintenance of industrial equipment extends its useful life, improves its efficiency, reduces the number of failures, and increases the safety of its use. This study proposes a methodology to develop a predictive maintenance tool based on infrared thermographic measures capable of anticipating failures in industrial equipment. The thermal response of selected equipment in normal operation and in controlled induced anomalous operation was analyzed. The characterization of these situations enabled the development of a machine learning system capable of predicting malfunctions. Different options within the available conventional machine learning techniques were analyzed, assessed, and finally selected for electronic equipment maintenance activities. This study provides advances towards the robust application of machine learning combined with infrared thermography and augmented reality for maintenance applications of industrial equipment. The predictive maintenance system finally selected enables automatic quick hand-held thermal inspections using 3D object detection and a pose estimation algorithm, making predictions with an accuracy of 94% at an inference time of 0.006 s. [ABSTRACT FROM AUTHOR]

Traditionally Caenorhabditis elegans lifespan assays are performed by manually inspecting nematodes with a dissection microscope, which involves daily counting of live/dead worms cultured in Petri plates for 21–25 days. This manual inspection requires the screening of hundreds of worms to ensure statistical robustness, and is therefore a time-consuming approach. In recent years, various automated artificial vision systems have been reported to increase the throughput, however they usually provide less accurate results than manual assays. The main problems identified when using these vision systems are the false positives and false negatives, which occur due to culture media changes, occluded zones, dirtiness or condensation of the Petri plates. In this work, we developed and described a new C. elegans monitoring machine, SiViS, which consists of a flexible and compact platform design to analyse C. elegans cultures using the standard Petri plates seeded with E. coli. Our system uses an active vision illumination technique and different image-processing pipelines for motion detection, both previously reported, providing a fully automated image processing pipeline. In addition, this study validated both these methods and the feasibility of the SiViS machine for lifespan experiments by comparing them with manual lifespan assays. Results demonstrated that the automated system yields consistent replicates (p-value log rank test 0.699), and there are no significant differences between automated system assays and traditionally manual assays (p-value 0.637). Finally, although we have focused on the use of SiViS in longevity assays, the system configuration is flexible and can, thus, be adapted to other C. elegans studies such as toxicity, mobility and behaviour. [ABSTRACT FROM AUTHOR]

Assistive technologies help all persons with disabilities to improve their accessibility in all aspects of their life. The AIDE European project contributes to the improvement of current assistive technologies by developing and testing a modular and adaptive multimodal interface customizable to the individual needs of people with disabilities. This paper describes the computer vision algorithms part of the multimodal interface developed inside the AIDE European project. The main contribution of this computer vision part is the integration with the robotic system and with the other sensory systems (electrooculography (EOG) and electroencephalography (EEG)). The technical achievements solved herein are the algorithm for the selection of objects using the gaze, and especially the state-of-the-art algorithm for the efficient detection and pose estimation of textureless objects. These algorithms were tested in real conditions, and were thoroughly evaluated both qualitatively and quantitatively. The experimental results of the object selection algorithm were excellent (object selection over 90%) in less than 12 s. The detection and pose estimation algorithms evaluated using the LINEMOD database were similar to the state-of-the-art method, and were the most computationally efficient. [ABSTRACT FROM AUTHOR]

Automated lifespan determination for C. elegans cultured in standard Petri dishes is challenging. Problems include occlusions of Petri dish edges, aggregation of worms, and accumulation of dirt (dust spots on lids) during assays, etc. This work presents a protocol for a lifespan assay, with two image-processing pipelines applied to different plate zones, and a new data post-processing method to solve the aforementioned problems. Specifically, certain steps in the culture protocol were taken to alleviate aggregation, occlusions, contamination, and condensation problems. This method is based on an active illumination system and facilitates automated image sequence analysis, does not need human threshold adjustments, and simplifies the techniques required to extract lifespan curves. In addition, two image-processing pipelines, applied to different plate zones, were employed for automated lifespan determination. The first image-processing pipeline was applied to a wall zone and used only pixel level information because worm size or shape features were unavailable in this zone. However, the second image-processing pipeline, applied to the plate centre, fused information at worm and pixel levels. Simple death event detection was used to automatically obtain lifespan curves from the image sequences that were captured once daily throughout the assay. Finally, a new post-processing method was applied to the extracted lifespan curves to filter errors. The experimental results showed that the errors in automated counting of live worms followed the Gaussian distribution with a mean of 2.91% and a standard deviation of ±12.73% per Petri plate. Post-processing reduced this error to 0.54 ± 8.18% per plate. The automated survival curve incurred an error of 4.62 ± 2.01%, while the post-process method reduced the lifespan curve error to approximately 2.24 ± 0.55%. [ABSTRACT FROM AUTHOR]

Background: The aging of the population and the progressive increase of life expectancy in developed countries is leading to a high incidence of age-related cerebrovascular diseases, which affect people's motor and cognitive capabilities and might result in the loss of arm and hand functions. Such conditions have a detrimental impact on people's quality of life. Assistive robots have been developed to help people with motor or cognitive disabilities to perform activities of daily living (ADLs) independently. Most of the robotic systems for assisting on ADLs proposed in the state of the art are mainly external manipulators and exoskeletal devices. The main objective of this study is to compare the performance of an hybrid EEG/EOG interface to perform ADLs when the user is controlling an exoskeleton rather than using an external manipulator. Methods: Ten impaired participants (5 males and 5 females, mean age 52 ± 16 years) were instructed to use both systems to perform a drinking task and a pouring task comprising multiple subtasks. For each device, two modes of operation were studied: synchronous mode (the user received a visual cue indicating the sub-tasks to be performed at each time) and asynchronous mode (the user started and finished each of the sub-tasks independently). Fluent control was assumed when the time for successful initializations ranged below 3 s and a reliable control in case it remained below 5 s. NASA-TLX questionnaire was used to evaluate the task workload. For the trials involving the use of the exoskeleton, a custom Likert-Scale questionnaire was used to evaluate the user's experience in terms of perceived comfort, safety, and reliability. Results: All participants were able to control both systems fluently and reliably. However, results suggest better performances of the exoskeleton over the external manipulator (75% successful initializations remain below 3 s in case of the exoskeleton and bellow 5s in case of the external manipulator). Conclusions: Although the results of our study in terms of fluency and reliability of EEG control suggest better performances of the exoskeleton over the external manipulator, such results cannot be considered conclusive, due to the heterogeneity of the population under test and the relatively limited number of participants. [ABSTRACT FROM AUTHOR]

A hyperspectral image analysis was used to characterize heat treatment in oat flour, performed by treating oat flour at 80, 100 and 130 °C for 30 min. Images from both original oat and treated flours were captured, and hyperspectral information was collected. Oat flours were used to obtain composite flours based on two different substitution levels (10 and 20%) of wheat flour. Composite breads were produced from the obtained flours. A battery of analyses was run to characterize them in terms of physical properties. The hyperspectral information of oat flours was analyzed by multivariate statistics and a pattern evolution-depending temperature was observed. Similarly, a set of the physical properties of breads was analyzed based on multivariate statistics, and a pattern of temperature-dependent evolution, in addition to the substitution level, was also recognized. Multivariate non linear regressions were done between both data sets to study their relationship and high values in the calibration and cross-validation results obtained. The changes undergone by oat flour during treatment were characterized with hyperspectral information, which could represent a non destructive monitoring tool to then regulate it until oat flours are obtained that confer composite bread adequate properties. [ABSTRACT FROM AUTHOR]

Some chia seed flour effects relating to different bread-making process phases and variables were studied by distinct image analysis and physicochemical techniques. Wheat flours with three different degrees of substitution (5%, 10% and 15%) were tested. In technological terms, the aim was to study the influence and properties of chia flour on several relevant parameters, such as pasting properties, growth kinetics and internal crumb structure during dough fermentation; and baking process, mass loss, water activity and texture profile of the end product during its storage. Some changes in pasting properties were observed. The effects obtained by image analysis techniques proved that addition of chia improved gas retention in dough and cut the time required to reach maximum dough development. A delay in hardness and water loss during storage of breads was also observed. Bread presented reduced water activity, and contained the same amount of moisture compared with the control. The mucilage provided by chia has properties that can explain these observed effects given the influence on water-holding capacity and its interactions with gluten proteins throughout the gluten matrix-forming process. [ABSTRACT FROM AUTHOR]

The objective of this study was to test the capability of a SW-NIR hyperspectral image technique to detect adulterations in wheat flour and bread with cheap grains, such us sorghum, oats and corn, and to compare the hyperspectral information with the physicochemical alterations in the properties of products. Wheat flour was adulterated at four different degrees (2.5, 5, 7.5 and 10%) with sorghum, oat and corn flours. Flours were prepared and used to make bread. Flours and breads were characterized according to several physicochemical parameters (pasting properties, water activity, mass loss during the baking process and texture profile analysis). Crumbs were extracted from breads and conditioned. Hyperspectral image captures were taken of both flours and conditioned crumbs. The data analysis was based on multivariate statistical process control method (MSPC), where the differentiation of adulterated samples was observed in all cases for both flours and crumbs. Finally, in order to relate the image analysis results and the adulterated sample properties, a correlation significance map was created between the physicochemical properties of samples and the multivariate statistical parameters. The SW-NIR image technique was capable of detecting adulterations in each case and high correlation significances were observed (α = 0.01) between wavelengths from specific spectra zones and the physicochemical properties of samples. [ABSTRACT FROM AUTHOR]