New method for a SEM-based quantitative microstructural clay analysis - MiCA
روش جدید برای تجزیه و تحلیل کمی خاک رس ریزساختاری مبتنی بر SEM - MiCA-2021
The soil microstructure is recognised to strongly influence the mechanical behaviour of both coarse and fine geomaterials. Proper identification and tracking of the shape and position of the particles has become more and more critical to form a link between the micro and macro behaviour. Scanning Electron Microscopy (SEM) has been widely used in the last decades to study the clay fabric variation with its mechanical behaviour and physical properties. However, the particles orientation has so far been discussed only from a qualitative point of view due to the lack of updated, automatised quantification tools. Consequently, developing constitutive models that correlate the soil micro structure to its mechanical behaviour is not feasible. In order to make a step forward in this direction, an Image Analysis based code called MiCA (microstructural clay analyser) capable of quantifying the particle orientation and the porosity of clay samples through the analysis of SEM micrographs was developed in this study. The code reliability was first validated through the application to geometrical reference patterns, then to textbook micrographs illustrating typical clay fabrics (dispersed, honeycomb, flocculated and aggre- gated), and finally to high quality images. MiCA showed good accuracy in the results obtained, regardless of the number of lines in the image, the complexity of the geometrical shapes and the pixel size of the analysed graphs. Therefore, MiCA can be considered suitable for quantitative analysis of the particle orientation and/or pores shape in clay materials.
Keywords: Image analysis | Computer vision | Quantitative analysis | Clay particles orientation | SEM | Clay porosity
Calibrated color measurement of cashmere using a novel computer vision system
اندازه گیری کالیبره رنگ ترمه با استفاده از یک سیستم بینایی ماشین جدید-2021
Color of cashmere fiber is a key component of textile quality inspection. However, the inspection of cashmere color was determined by human vision system, which was less accuracy and time-consuming. Computer vision system (CVS) is considered as a promising technique to objectively and precisely test color. In the present work, a novel color measurement system for cashmere was proposed. Totally 29 cashmere samples with different color were adopted as standard samples to calibrate color conversion model. The correlation coefficient of L, a, b values between the two systems was separately calculated high to 0.99, 0.96 and 0.93 for the whole samples. The proposed method was further validated by other 15 samples, indicating the accuracy of the novel CVS. Besides, due to the high accuracy and strong representativeness of the new method, the categories of cashmere, which were normally tested by subjective visual assessment, could be determined by the present results.
Keywords: Cashmere color | Computer vision | Image analysis | Calibration model | Classification
Classification of fermented cocoa beans (cut test) using computer vision
طبقه بندی دانه های کاکائو تخمیر شده (تست برش) با استفاده از بینایی ماشین-2021
Fermentation of cocoa beans is a critical step for chocolate manufacturing, since fermentation influences the development of flavour, affecting components such as free amino acids, peptides and sugars. The degree of fermentation is determined by visual inspection of changes in the internal colour and texture of beans, through the cut-test. Although considered standard for evaluation of fermentation in cocoa beans, this method is time consuming and relies on specialized personnel. Therefore, this study aims to classify fermented cocoa beans using computer vision as a fast and accurate method. Imaging and image analysis provides hand-crafted features computed from the beans, that were used as predictors in random decision forests to classify the samples. A total of 1800 beans were classified into four grades of fermentation. Concerning all image features, 0.93 of accuracy was obtained for validation of unbalanced dataset, with precision of 0.85, recall of 0.81. Although the unbalanced dataset represents actual variation of fermentation, the method was tested for a balanced dataset, to investigate the influence of a smaller number of samples per class, obtaining 0.92, 0.92 and 0.90 for accuracy, precision and recall, respectively. The technique can evolve into an industrial application with a proper integration framework, substituting the traditional method to classify fermented cocoa beans.
Keywords: Chocolate | Cut-test | Food quality | Analytical method | Image analysis | Random decision forest
A deep learning approach to the screening of malaria infection: Automated and rapid cell counting, object detection and instance segmentation using Mask R-CNN
یک روش یادگیری عمیق برای غربالگری عفونت مالاریا: شمارش خودکار و سریع سلول ها ، تشخیص اشیاء و تقسیم بندی نمونه با استفاده از Mask R-CNN-2021
Accurate and early diagnosis is critical to proper malaria treatment and hence death prevention. Several com- puter vision technologies have emerged in recent years as alternatives to traditional microscopy and rapid diagnostic tests. In this work, we used a deep learning model called Mask R-CNN that is trained on uninfected and Plasmodium falciparum-infected red blood cells. Our predictive model produced reports at a rate 15 times faster than manual counting without compromising on accuracy. Another unique feature of our model is its ability to generate segmentation masks on top of bounding box classifications for immediate visualization, making it superior to existing models. Furthermore, with greater standardization, it holds much potential to reduce errors arising from manual counting and save a significant amount of human resources, time, and cost.
Keywords: Malaria diagnosis | Mask R-CNN | Computer vision | Image analysis
Defect detection and quantification in electroluminescence images of solar PV modules using U-net semantic segmentation
تشخیص و تعیین کمبود در تصاویر الکترولومینسانس ماژول های PV خورشیدی با استفاده از تقسیم بندی معنایی U-net-2021
Electroluminescence (EL) images enable defect detection in solar photovoltaic (PV) modules that are otherwise invisible to the naked eye, much the same way an x-ray enables a doctor to detect cracks and fractures in bones. The prevalence of multiple defects, e.g. micro cracks, inactive regions, gridline defects, and material defects, in PV module can be quantiﬁed with an EL image. Modern, deep learning tech- niques for computer vision can be applied to extract the useful information contained in the images on entire batches of PV modules. Defect detection and quantiﬁcation in EL images can improve the efﬁciency and the reliability of PV modules both at the factory by identifying potential process issues and at the PV plant by identifying and reducing the number of faulty modules installed. In this work, we train and test a semantic segmentation model based on the u-net architecture for EL image analysis of PV modules made from mono-crystalline and multi-crystalline silicon wafer-based solar cells. This work is focused on developing and testing a deep learning method for computer vision that is independent of the equipment used to generate the EL images, independent of the wafer-based module design, and independent of the image quality.© 2021 Elsevier Ltd. All rights reserved.
Keywords: Electroluminescence | EL | PV | U-net | Semantic segmentation | Machine learning
Unrest index for estimating thermal comfort of poultry birds (Gallus gallus domesticus) using computer vision techniques
شاخص ناآرامی برای برآورد آسایش حرارتی پرندگان طیور (Gallus gallus domesticus) با استفاده از تکنیک های بینایی ماشین-2021
Behaviour can be used to infer animal welfare states. Poultry birds tend to move less under conditions of thermal stress; hence the hypothesis of this research is that this unrest behaviour can be used as an indicator of thermal comfort. The objective was to develop an Unrest Index for poultry bird’s sensitive to changes in this behaviour under different air temperature conditions. The proposed Unrest Index was based on the Hausdorff distance measure and was tested on recorded videos of laying hens and broilers breeders, obtained in different experiments. The index was efficient in detecting the unrest of poultry birds in different thermal conditions and, in conditions above thermoneutrality, the birds moved significantly less. The distribution Unrest Index of data for each thermal condition tested was shown to be asymmetric. However, there seems to be a tendency to reverse this asymmetry when the conditions are thermal comfort and heat stress. It is suggested that the Unrest Index can be used to estimate the thermal comfort of poultry birds and that further studies on the asymmetry of the index data should be carried out in order to identify of the thermoneutrality zone of birds in a non-invasive way. The Unrest Index and the computer vision techniques adopted to assess poultry thermal comfort automatically were efficient in demonstrating differences in bird agitation in distinct thermal stress conditions. The low computational effort and the mathematical simplicity of the model allows the Unrest Index to incorporate bird surveillance systems and estimate thermal comfort automatically.
Keywords: Animal behaviour | Hausdorff distance | Comfort index | Precision livestock farming | Image analysis
Monitoring the hot-air drying process of organically grown apples (cv: Gala) using computer vision
نظارت بر فرایند خشک شدن سیب های ارگانیک در هوا (cv: Gala) با استفاده از بینایی ماشین-2021
The feasibility of using a computer-vision (CV) system embedded in a hot-air dryer for nondestructive and real-time monitoring of the drying behaviour of organic apples was investigated in the present study. Apple cylinders were subjected to anti-browning treatments with different dipping solutions (water as control, trehalose (4% w/v) or trehalose þ ascorbic acid (4% w/v and 1% w/v)) and dipping pressures (atmospheric, 101.3 kPa and sub-atmospheric pressure, 50 kPa) followed by drying at 60 C to a final dry basis moisture content of 0.18 g g1. The CV system was used as an in-line process analytical technology (PAT) tool to capture images reflecting the physico-chemical changes during product drying coupled with in-line mass changes and off-line reference analyses. The spatial and colour changes from the image analysis described well the complex and non-homogenous nature of apple drying. The results of spatial changes allowed successful development of accurate linear prediction models for moisture content as a function of area shrinkage (on scaled variables) with excellent prediction capability (|BIAS| < 8.5 103, RMSE < 0.04, Adj-R2 ~ ¼ 99%). Also, the CV system identified the differently pre-treated samples, particularly the dipping pressures as reflected by two linear models and their respective parameters. The obtained results demonstrate the versatile advantages of CV systems as an in-line tool for continuous, real-time monitoring of apples during drying. The insights from this study can provide a platform for applications of CV embedded ‘Smart dryers’ as an efficient monitoring and control system for industrial drying processes.
Keywords: Malus domestica B. | Image analysis | Dipping treatments | Vacuum impregnation | Smart drying | Area shrinkage
Phase volume quantification of agarose-ghee gels using 3D confocal laser scanning microscopy and blending law analysis: A comparison
اندازه گیری حجم فاز ژل های agarose-ghee با استفاده از میکروسکوپ اسکن لیزر کانفوکال 3D و تجزیه و تحلیل قانون: یک مقایسه -2020
A thorough understanding of the phase behaviour of biomaterial composites is imperative for manipulating the structural and textural properties in novel food products. This study probed the phase behaviour of a model system comprising agarose and a varying concentration of ghee. Results obtained from scanning electron microscopy (SEM), micro differential scanning calorimetry (DSC), Fourier transform infrared spectroscopy (FTIR) and dynamic oscillation in-shear revealed discontinuous and hard inclusions of ghee reinforcing the continuous, weaker agarose matrix with increasing concentrations of the former. Phase behaviour of the system was quantified in parallel with a novel method combining 3D confocal laser scanning microscopy (CLSM) imaging and image analysis software - FIJI and Imaris - in an effort to substantiate the efficacy of the microscopic protocol in quantifying phase behaviour. Phase volumes recorded with the microscopic protocol were in close agreement to those modelled with the Lewis-Nielsen blending law using small-deformation dynamic oscillation. However, results indicated that the inner filtering effect or ‘self-shadowing’ observed commonly in CLSM images may pose a limitation to the application of this technique, necessitating further development before it can be applied to more complex, industrially relevant systems.
Keywords: Lewis-Nielsen blending law | Phase behaviour | Confocal laser scanning microscopy | 3D imaging | Image analysis
AI-PLAX: AI-based placental assessment and examination usingphotos
AI-PLAX: ارزیابی و معاینه جفت مبتنی بر هوش مصنوعی با استفاده از عکس-2020
Post-delivery analysis of the placenta is useful for evaluating health risks of both the mother and baby. In the U.S., however, only about 20% of placentas are assessed by pathology exams, and placental data is often missed in pregnancy research because of the additional time, cost, and expertise needed. A computer-based tool that can be used in any delivery setting at the time of birth to provide an immediate and comprehensive placental assessment would have the potential to not only to improve health care, but also to radically improve medical knowledge. In this paper, we tackle the problem of automatic placental assessment and examination using photos. More concretely, we first address morphological characterization, which includes the tasks of placental image segmentation, umbilical cord insertion point localization, and maternal/fetal side classification. We also tackle clinically meaningful feature analysis of placentas, which comprises detection of retained placenta (i.e., incomplete placenta), umbilical cord knot, meconium, abruption, chorioamnionitis, and hypercoiled cord, and categorization of umbilical cord insertion type. We curated a dataset consisting of approximately 1300 placenta images taken at Northwestern Memorial Hospital, with hand-labeled pixel-level segmentation map, cord insertion point and other information extracted from the associated pathology reports. We developed the AI-based Placental Assessment and Examination system (AI-PLAX), which is a novel two-stage photograph-based pipeline for fully automated analysis. In the first stage, we use three encoder-decoder convolutional neural networks with a shared encoder to address morphological characterization tasks by employing a transfer-learning training strategy. In the second stage, we employ distinct sub-models to solve different feature analysis tasks by using both the photograph and the output of the first stage. We evaluated the effectiveness of our pipeline by using the curated dataset as well as the pathology reports in the medical record. Through extensive experiments, we demonstrate our system is able to produce accurate morphological characterization and very promising performance on aforementioned feature analysis tasks, all of which may possess clinical impact and contribute to future pregnancy research. This work is the first for comprehensive, automated, computer-based placental analysis and will serve as a launchpad for potentially multiple future innovations.
Keywords: Deep learning | Transfer learning | Placenta | Photo image analysis | Pathology
Optimal surface estimation and thresholding of confocal microscope images of biofilms using Beers Law
تخمین بهینه سطح و آستانه گرفتن تصاویر میکروسکوپ کانفوکال بیوفیلم ها با استفاده از قانون Beers -2020
Beers Law explains how light attenuates into thick specimens, including thick biofilms. We use a Bayesian optimality criterion, the maximum of the posterior probability distribution, and computationally efficiently fit Beers Law to the 3D intensity data collected from thick living biofilms by a confocal scanning laser microscope. Using this approach the top surface of the biofilm and an optimal image threshold can be estimated. Biofilm characteristics, such as bio-volumes, can be calculated from this surface. Results from the Bayesian approach are compared to other approaches including the method of maximum likelihood or simply counting bright pixels. Uncertainty quantification (i.e., error bars) can be provided for the parameters of interest. This approach is applied to confocal images of stained biofilms of a common lab strain of Pseudomonas aeruginosa, stained biofilms of Janthinobacterium isolated from the Antarctic, and biofilms of Staphylococcus aureus that have been genetically modified to fluoresce green.
Keywords: Attenuation | Thresholding | Maximum likelihood | Beer-Lambert Law | Bayesian | Confocal microscope image analysis