Abstract: Estimating the 6D pose of objects is a core step for robot bin-picking tasks. The problem is that various objects are usually randomly stacked with heavy occlusion in real applications. In this work, we propose a method to regress 6D poses by predicting three points for each object in the 3D point cloud through deep learning. To solve the ambiguity of symmetric pose, we propose a labeling method to help the network converge better. Based on the predicted pose, an iterative method is employed for pose optimization. In real-world experiments, our method outperforms the classical approach in both precision and recall.
Abstract: Engagement is one of the most important factors in determining successful outcomes and deep learning in students. Existing approaches to detect student engagement involve periodic human observations that are subject to inter-rater reliability. Our solution uses real-time multimodal multisensor data labeled by objective performance outcomes to infer the engagement of students. The study involves four students with a combined diagnosis of cerebral palsy and a learning disability who took part in a 3-month trial over 59 sessions. Multimodal multisensor data were collected while they participated in a continuous performance test. Eye gaze, electroencephalogram, body pose, and interaction data were used to create a model of student engagement through objective labeling from the continuous performance test outcomes. In order to achieve this, a type of continuous performance test is introduced, the Seek-X type. Nine features were extracted including high-level handpicked compound features. Using leave-one-out cross-validation, a series of different machine learning approaches were evaluated. Overall, the random forest classification approach achieved the best classification results. Using random forest, 93.3% classification for engagement and 42.9% accuracy for disengagement were achieved. We compared these results to outcomes from different models: AdaBoost, decision tree, k-Nearest Neighbor, naïve Bayes, neural network, and support vector machine. We showed that using a multisensor approach achieved higher accuracy than using features from any reduced set of sensors. We found that using high-level handpicked features can improve the classification accuracy in every sensor mode. Our approach is robust to both sensor fallout and occlusions. The single most important sensor feature to the classification of engagement and distraction was shown to be eye gaze. It has been shown that we can accurately predict the level of engagement of students with learning disabilities in a real-time approach that is not subject to inter-rater reliability, human observation or reliant on a single mode of sensor input. This will help teachers design interventions for a heterogeneous group of students, where teachers cannot possibly attend to each of their individual needs. Our approach can be used to identify those with the greatest learning challenges so that all students are supported to reach their full potential.
Abstract: In this study, an automated building and tree detection
method is proposed using DSM data and true orthophoto image. A
multiscale matched filtering is used on DSM data. Therefore, first
watershed transform is applied. Then, Otsu’s thresholding method
is used as an adaptive threshold to segment each watershed region.
Detected objects are masked with NDVI to separate buildings and
trees. The proposed method is able to detect buildings and trees
without entering any elevation threshold. We tested our method on
ISPRS semantic labeling dataset and obtained promising results.
Abstract: Chitosan is a natural polysaccharide prepared by the N-deacetylation of chitin. In this study, the physicochemical and antibacterial properties of chitosan nanoparticles, produced by ultrasound irradiation, were evaluated. The physicochemical properties of the nanoparticles were determined by dynamic light scattering and zeta potential analysis. Chitosan nanoparticles inhibited the growth of E. coli. The minimum inhibitory concentration (MIC) values were lower than 0.5 mg/mL, and the minimum bactericidal concentration (MBC) values were similar or higher than MIC values. Confocal laser scanning micrographs (CLSM) were used to observe the interaction between E. coli suspensions mixed with FITC-labeled chitosan polymers and nanoparticles.
Abstract: The paper studies the role of food labeling in order to promote healthy eating issue in Georgia. The main focus of the research is directed to consumer attitudes regarding food labeling. The methodology of the paper is based on the focus group work, as well as online and face to face surveys. The data analysis has been provided through ANOVA. The study proves that the impact of variables such as the interest, awareness, reliability, assurance and satisfaction of consumers' on buying decision, is statistically important. The study reveals that consumers’ perception regarding to food labeling is positive, but their level of knowledge and ability is rather low. It is urgent to strengthen marketing promotions strategies in the process of implementations of food security policy in Georgia.
Abstract: In this paper, we propose labeling based RANSAC algorithm for lane detection. Advanced driver assistance systems (ADAS) have been widely researched to avoid unexpected accidents. Lane detection is a necessary system to assist keeping lane and lane departure prevention. The proposed vision based lane detection method applies Canny edge detection, inverse perspective mapping (IPM), K-means algorithm, mathematical morphology operations and 8 connected-component labeling. Next, random samples are selected from each labeling region for RANSAC. The sampling method selects the points of lane with a high probability. Finally, lane parameters of straight line or curve equations are estimated. Through the simulations tested on video recorded at daytime and nighttime, we show that the proposed method has better performance than the existing RANSAC algorithm in various environments.
Abstract: Muscle regeneration after injury (as irradiation) is of great importance. However, the molecular and cellular mechanisms are still unclear. Cytokines are believed to play fundamental role in the different stages of muscle regeneration. They are secreted by many cell populations, but the predominant producers are macrophages and helper T cells. On the other hand, it has been shown that adipose tissue derived stromal/stem cell (ASC) injection could improve muscle regeneration. Stem cells probably induce the coordinated modulations of gene expression in different macrophage cells. Therefore, we investigated the patterns and timing of changes in gene expression of different cytokines occurring upon stem cells loading. Muscle regeneration was studied in an irradiated muscle of minipig animal model in presence or absence of ASC treatment (irradiated and treated with ASCs, IRR+ASC; irradiated not-treated with ASCs, IRR; and non-irradiated no-IRR). We characterized macrophage populations by immunolabeling in the different conditions. In our study, we found mostly M2 and a few M1 macrophages in the IRR+ASC samples. However, only few M2b macrophages were noticed in the IRR muscles. In addition, we found intensive fibrosis in the IRR samples. With in situ hybridization and immunolabeling, we analyzed the cytokine expression of the different macrophages and we showed that M2d macrophage are the most abundant in the IRR+ASC samples. By in situ hybridization, strong expression of the transforming growth factor β (TGF-β) was observed in the IRR+ASC but very week in the IRR samples. But when we analyzed TGF-β level with immunolabeling the expression was very different: many M2 macrophages showed week expression in IRR+ASC and few cells expressing stronger level in IRR muscles. Therefore, we investigated the MMP expressions in the different muscles. Our data showed that the M2 macrophages of the IRR+ASC muscle expressed MMP2 proteins. Our working hypothesis is that MMP2 expression of the M2 macrophages can decrease fibrosis in the IRR+ASC muscle by capturing TGF-β.
Abstract: Music has always been an integral part of human’s daily lives. But, for the most people, reading musical score and turning it into melody is not easy. This study aims to develop an Automatic music score recognition system using digital image processing, which can be used to read and analyze musical score images automatically. The technical approaches included: (1) staff region segmentation; (2) image preprocessing; (3) note recognition; and (4) accidental and rest recognition. Digital image processing techniques (e.g., horizontal /vertical projections, connected component labeling, morphological processing, template matching, etc.) were applied according to musical notes, accidents, and rests in staff notations. Preliminary results showed that our system could achieve detection and recognition rates of 96.3% and 91.7%, respectively. In conclusion, we presented an effective automated musical score recognition system that could be integrated in a system with a media player to play music/songs given input images of musical score. Ultimately, this system could also be incorporated in applications for mobile devices as a learning tool, such that a music player could learn to play music/songs.
Abstract: Hyperspectral imagery (HSI) typically provides a
wealth of information captured in a wide range of the
electromagnetic spectrum for each pixel in the image. Hence, a
pixel in HSI is a high-dimensional vector of intensities with a
large spectral range and a high spectral resolution. Therefore, the
semantic interpretation is a challenging task of HSI analysis. We
focused in this paper on object classification as HSI semantic
interpretation. However, HSI classification still faces some issues,
among which are the following: The spatial variability of spectral
signatures, the high number of spectral bands, and the high cost
of true sample labeling. Therefore, the high number of spectral
bands and the low number of training samples pose the problem of
the curse of dimensionality. In order to resolve this problem, we
propose to introduce the process of dimensionality reduction trying
to improve the classification of HSI. The presented approach is a
semi-supervised band selection method based on spatial hypergraph
embedding model to represent higher order relationships with
different weights of the spatial neighbors corresponding to the
centroid of pixel. This semi-supervised band selection has been
developed to select useful bands for object classification. The
presented approach is evaluated on AVIRIS and ROSIS HSIs
and compared to other dimensionality reduction methods. The
experimental results demonstrate the efficacy of our approach
compared to many existing dimensionality reduction methods for
HSI classification.
Abstract: The food industry nowadays is becoming customer-oriented and needs faster response time to deal with food incidents. There is a deep need for good traceability systems to help the supply chain (SC) partners to minimize production and distribution of unsafe or poor quality products, which in turn will enhance the food SC performance. The current food labeling systems implemented in developing countries cannot guarantee that food is authentic, safe and of good quality. Therefore, the use of origin labels, mainly the geographical indications (GIs), allows SC partners to define quality standards and defend their products' reputation. According to our knowledge there are no studies discussed the use of GIs in developing countries. This research represents a research schema about the implementation of European quality labeling system in developing countries and its impact on enhancing SC performance. An empirical study was conducted on the Egyptian traditional food sector based on a sample of seven restaurants implementing the Med-diet labeling system. First, in-depth interviews were carried out to analyze the Egyptian traditional food SC. Then, a framework was developed to link the European quality labeling system and SC performance. Finally, a structured survey was conducted based on the applied framework to investigate the impact of Med-diet labeling system on the SC performance. The research provides an applied framework linking Med-diet quality labeling system to SC performance of traditional food sector in developing countries generally and especially in the Egyptian traditional food sector. The framework can be used as a SC performance management tool to increase the effectiveness and efficiency of food industry's SC performance.
Abstract: The use of QR (Quick Response Codes) codes for customer scanning with mobile phones is a rapidly growing trend. The QR code can provide the consumer with product information, user guides, product use, competitive pricing, etc. One sector for QR use has been in retail, through the use of Electronic Shelf Labeling (henceforth, ESL). In Europe, the use of ESL for pricing has been in practice for a number of years but continues to lag in acceptance in North America. Stated concerns include costs as a key constraint, but there is also evidence that consumer acceptance represents a limitation as well. The purpose of this study is to present the findings of a consumer based study to gage the impact on their use in the retail food sector.
Abstract: Effective statistical feature extraction and classification are important in image-based automatic inspection and analysis. An automatic wood species recognition system is designed to perform wood inspection at custom checkpoints to avoid mislabeling of timber which will results to loss of income to the timber industry. The system focuses on analyzing the statistical pores properties of the wood images. This paper proposed a fuzzy-based feature extractor which mimics the experts’ knowledge on wood texture to extract the properties of pores distribution from the wood surface texture. The proposed feature extractor consists of two steps namely pores extraction and fuzzy pores management. The total number of statistical features extracted from each wood image is 38 features. Then, a backpropagation neural network is used to classify the wood species based on the statistical features. A comprehensive set of experiments on a database composed of 5200 macroscopic images from 52 tropical wood species was used to evaluate the performance of the proposed feature extractor. The advantage of the proposed feature extraction technique is that it mimics the experts’ interpretation on wood texture which allows human involvement when analyzing the wood texture. Experimental results show the efficiency of the proposed method.
Abstract: The aims of this research are to broaden the study on the relationship between emotional intelligence and counterproductive work behavior (CWB). The study sample consisted in 441 Romanian employees from companies all over the country. Data has been collected through web surveys and processed with SPSS. The results indicated an average correlation between the two constructs and their sub variables, employees with a high level of emotional intelligence tend to be less aggressive. In addition, labeling was considered an individual difference which has the power to influence the level of employee aggression. A regression model was used to underline the importance of emotional intelligence together with labeling as predictors of CWB. Results have shown that this regression model enforces the assumption that labeling and emotional intelligence, taken together, predict CWB. Employees, who label themselves as victims and have a low degree of emotional intelligence, have a higher level of CWB.
Abstract: We consider the problem of placing labels of the points
on a plane. For each point, its position, the size of its label and a
priority are given. Moreover, several candidates of its label positions
are prespecified, and each of such label positions is assigned a
priority. The objective of our problem is to maximize the total sum
of priorities of placed labels and their points. By refining a labeling
algorithm that can use these priorities, we propose a new heuristic
algorithm which is more suitable for treating the assigned priorities.
Abstract: We developed new processes which can collect and detect rapidly airborne pathogens such as the avian flu virus for the pandemic prevention. The fluorescence antibody technique is known as one of high-sensitive detection methods for viruses, but this needs up to a few hours to bind sufficient fluorescence dyes to viruses for detection. In this paper, we developed a mist-labeling can detect substitution viruses in a short time to improve the binding rate of fluorescent dyes and substitution viruses by the micro reaction process. Moreover, we developed the rapid detection system with the above “mist labeling”. The detection system set with a sampling bag collecting patient’s breath and a cartridge can detect automatically pathogens within 10 minutes.
Abstract: Breast region segmentation is an essential prerequisite in computerized analysis of mammograms. It aims at separating the breast tissue from the background of the mammogram and it includes two independent segmentations. The first segments the background region which usually contains annotations, labels and frames from the whole breast region, while the second removes the pectoral muscle portion (present in Medio Lateral Oblique (MLO) views) from the rest of the breast tissue. In this paper we propose hybridization of Connected Component Labeling (CCL), Fuzzy, and Straight line methods. Our proposed methods worked good for separating pectoral region. After removal pectoral muscle from the mammogram, further processing is confined to the breast region alone. To demonstrate the validity of our segmentation algorithm, it is extensively tested using over 322 mammographic images from the Mammographic Image Analysis Society (MIAS) database. The segmentation results were evaluated using a Mean Absolute Error (MAE), Hausdroff Distance (HD), Probabilistic Rand Index (PRI), Local Consistency Error (LCE) and Tanimoto Coefficient (TC). The hybridization of fuzzy with straight line method is given more than 96% of the curve segmentations to be adequate or better. In addition a comparison with similar approaches from the state of the art has been given, obtaining slightly improved results. Experimental results demonstrate the effectiveness of the proposed approach.
Abstract: Laser projection or laser footprint detection is today widely used in many fields of robotics, measurement or electronics. The system accuracy strictly depends on precise laser footprint detection on target objects. This article deals with the laser line detection based on the RGB segmentation and the component labeling. As a measurement device was used the developed optical rangefinder. The optical rangefinder is equipped with vertical sweeping of the laser beam and high quality camera. This system was developed mainly for automatic exploration and mapping of unknown spaces. In the first section is presented a new detection algorithm. In the second section are presented measurements results. The measurements were performed in variable light conditions in interiors. The last part of the article present achieved results and their differences between day and night measurements.
Abstract: Support vector clustering (SVC) is an important kernelbased clustering algorithm in multi applications. It has got two main bottle necks, the high computation price and labeling piece. In this paper, we presented a modified SVC method, named Grid–SVC, to improve the original algorithm computationally. First we normalized and then we parted the interval, where the SVC is processing, using a novel Grid–based clustering algorithm. The algorithm parts the intervals, based on the density function of the data set and then applying the cartesian multiply makes multi-dimensional grids. Eliminating many outliers and noise in the preprocess, we apply an improved SVC method to each parted grid in a parallel way. The experimental results show both improvement in time complexity order and the accuracy.
Abstract: In this paper, a simple moving human detection method is proposed for video surveillance system or access monitoring system. The frame difference and noise threshold are used for initial detection of a moving human-object, and simple labeling method is applied for final human-object segmentation. The simulated results show that the applied algorithm is fast to detect the moving human-objects by performing 95% of correct detection rate. The proposed algorithm has confirmed that can be used as an intelligent video access monitoring system.
Abstract: We present a hardware oriented method for real-time
measurements of object-s position in video. The targeted application
area is light spots used as references for robotic navigation. Different
algorithms for dynamic thresholding are explored in combination
with component labeling and Center Of Gravity (COG) for highest
possible precision versus Signal-to-Noise Ratio (SNR). This method
was developed with a low hardware cost in focus having only one
convolution operation required for preprocessing of data.