Recognition Stage Research Articles

Joint non-negative and fuzzy coding with graph regularization for efficient data clustering

Non-negative matrix factorization (NMF) is an effective model in converting data into non-negative coefficient representation whose discriminative ability is usually enhanced to be used for diverse pattern recognition tasks. In NMF-based clustering, we often need to perform K-means on the learned coefficient as postprocessing step to get the final cluster assignments. This breaks the connection between the feature learning and recognition stages. In this paper, we propose to learn the non-negative coefficient matrix based on which we jointly perform fuzzy clustering, by viewing that each column of the dictionary matrix as a concept of each cluster. As a result, we formulate a new fuzzy clustering model, termed Joint Non-negative and Fuzzy Coding with Graph regularization (G-JNFC), and design an effective optimization method to solve it under the alternating direction optimization framework. Besides the convergence and computational complexity analysis on G-JNFC, we conduct extensive experiments on both synthetic and representative benchmark data sets. The results show that the proposed G-JNFC model is effective in data clustering.

Normas internacionales de información financiera y reconocimiento contable de jugadores de fútbol en clubes deportivos

La presente investigación tiene por objetivo diseñar una guía de procedimientos contables para el reconocimiento y medición de jugadores de fútbol en clubes deportivos del Ecuador. Para lograrlo, se ha empleado un enfoque cualitativo de indagación, además, una entrevista al Club Deportivo Cuenca y una ficha de revisión documental que han aportado significativamente a este trabajo y su propuesta. Los resultados de la investigación indican que los clubes no reflejan sus activos intangibles en sus estados financieros. En base a metodologías internacionales y lineamientos de la NIC 38 se ha establecido una guía para el tratamiento contable de jugadores de fútbol en sus etapas de reconocimiento y medición. El primer paso sería la revalorización de su plantilla actual, y seguir todos los procesos establecidos en este trabajo para lograr la transparencia, razonabilidad y comparabilidad que son características esenciales en la contabilidad.

Research on automatic generation technology of ejector pin for injection mold

When having narrow slot features in the mold core, the molded products cannot be smoothly removed from the mold core. Through ejector pin design, during the ejection process, the ejector pin can eject products to successfully leave the mold core. Under the framework of CAD software, this study developed the narrow slot feature recognition, integrated the mold design, and automated ejector pin production. Based on the geometric relationship of the surfaces, edges of the core’s narrow slot, this study summarized an algorithm to automatically recognize the narrow slot feature by programming an algorithm. Through computer programming calculation and judgment, the proposed method can automatically recognize narrow slot features to reduce the time at the recognition stage and record the diameters of the ejector pin and its coordinate position in feature parameters. At the mold design stage, through the acquisition of its parameters, it can automatically produce a suitable ejector pin. Through standardized design process, it can greatly shorten the design time to guide users to quickly complete the design, while reducing design errors to improve working efficiency by the automatic integration function.

Simplification of CAD Models by Automatic Recognition and Suppression of Blend Chains

This paper presents a CAD model simplification procedure that consists in recognizing and suppressing blend chains of certain types. The proposed method involves Euler operators (KEV, KEF, and KFMV) developed on top of an open-source geometric modeling kernel. The simplification process consists of two stages: recognition and suppression. The suppression stage ensures the geometric and topological validity of the simplification result. The proposed approach is targeted for use in batch mode, which poses strict requirements to the robustness of the suppression algorithm. The essential properties of the approach are its sustainability, predictability of the result, and extensible architecture, which allows new topological cases to be added without modifying the algorithm’s core. At the recognition stage, the algorithm constructs an attributed adjacency graph, which is then enriched with the information about types of edges, their properties, and assumed kinds of blend faces. At the suppression stage, the algorithm iterates through the adjacency graph and composes candidate blend faces into chains. For each face in a chain, local topology analysis is carried out to determine the corresponding sequence of Euler operators that are supposed to eliminate that face. The algorithm can be extended by adding descriptors of new topological cases into the processing. Upon applying the Euler operators, the affected edges are reconstructed to obtain a geometrically correct boundary representation of the model.

In silico design of a universal carrier protein: A potential solution to the variable immunogenicity observed in past vaccines against drugs of abuse

Abstract Vaccines against drugs of abuse (VADAs) show promise as a possible therapeutic for the alleviation of addiction due to their ability to modulate intake behavior via the altering of drug pharmacokinetics. In the past, however, these vaccines have failed to show clinical efficacy due to extensive variability in the magnitude of antibody response between individual vaccinees. Correspondingly, the future success of VADAs hinges directly on whether or not researchers can find ways to engineer more broadly effective vaccines. Differences in response must, to some extent, be the result of variable capacities for each individual’s immune system to process and/or present antigens. As such, we speculate that future VADAs will need to target the three key stages of vaccine processing, recognition, processing, and presentation, in order to achieve clinical effectiveness. Here, we describe a project aiming to engineer a broadly effective carrier protein based on this concept. Using haplotype frequency data provided by Be the Match, HLA-DQ and -DR haplotypes covering 99% of the population were selected for analysis. Next, the most common carrier proteins used in conjugate vaccine formulations were analyzed for MHC II epitope content using these haplotypes and NetMHCIIPan 3.2 prediction software. Finally, a modified version of the prediction output was used to excise the most likely MHC II epitopes. The highest-ranked epitopes were then stitched together using linkers designed with chemical conjugation sites and cathepsin cleavage sequences and subsequently tethered to cholera toxin subunit B. These results present the successful design of an in silico derived, broadly effective carrier protein to be used in future VADA formulations.

A Study on Pattern Recognition with the Histograms of Oriented Gradients in Distorted and Noisy Images

Histograms of oriented gradients (HOG) are still one of the most frequently used low-level features for pattern recognition in images. Despite their great popularity and simple implementation performance of the HOG features almost always has been measured on relatively high quality data which are far from real conditions. To fill this gap we experimentally evaluate their performance in the more realistic conditions, based on images affected by different types of noise, such as Gaussian, quantization, and salt-and-pepper, as well on images distorted by occlusions. Different noise scenarios were tested such anti-distortions during training as well as application of a proper denoising method in the recognition stage. As underpinned with experimental results, the negative impact of distortions and noise on object recognition with HOG features can be significantly reduced by employment of a proper denoising strategy.

Analysis of influencing factors in pre-evacuation time using Interpretive Structural Modeling

The significant reason for the delay in Required Safety Egress Time is the pre-evacuation behavior under building fire. The different behaviors after occupant received fire alarm or cues are influenced by numerous factors. Firstly, there are fifteen pre-evacuation influencing factors which are identified based on literature review and survey, including health status, social influence, affiliation, etc. The influencing factors on pre-evacuation behavior are further analyzed and explained. Secondly, the Interpretive Structural Modeling (ISM) is established and calculated. Then, the adjacent matrix and reachability matrix are acquired. Finally, a hierarchical structure of influencing factors bases on ISM is obtained. The results show that the structural model which presents the influencing factors of pre-evacuation behavior is a five levels hierarchical structure. The direct influencing factors include: actions at recognition & response stage, social influence, etc. The deep influencing factors which are also called origin factors include: age and gender. Therefore, the results indicate that root factors include age, gender and education level. These factors should be paid more attention in order to decrease the delay in the pre-evacuation time. Different evacuation measures should be seriously taken at different places with different crowds in order to increase the success rate of occupant evacuation.

RETRACTED ARTICLE: An efficient human face verification system based on ELBP: a high precision feature

Human Face Verification (HFV) is used for distinguishing the human as an authorized or unauthorized Person in security and biometric related applications. HFV has huge applications in many areas. HFV process involves the collection, detection, pre-processing, extraction, classifying and Verification stages. The main problem in HFV system is to detect the faces automatically from front view with variation in illumination, expression, and occlusion and so on. This work is mainly focused on extraction and recognition stages. LBP is widely used method for texture analysis. LBP usually compares the center pixel to its neighbors over a fixed radius. LBP is generally sensitive to noise. In order to overcome this problem ELBP method had been proposed which deals with the rotation invariant of face images. The proposed method is evaluated for noise environment, pose variations and emotions. For each one, the experimental results are tabulated and analyzed. From the result analysis, the proposed method provides better accuracy than existing methods.

An improved gaussian mixture hidden conditional random fields model for audio-based emotions classification

The analysis of human emotions plays a significant role in providing sufficient information about patients in monitoring their feelings for better management of their diseases. Audio-based emotions recognition has become a fascinating research interest for such domains during the last decade. Mostly, audio-based emotions systems depend on the recognition stage. The existing model has a common issue called objectivity suppositions problem, which might decrease the recognition rate. Therefore, this study investigates the improved version of a classifier that is based on hidden conditional random fields (HCRFs) model to classify emotional speech. In this model, we introduced a novel methodology that will incorporate multifaceted dissemination with the help of employing a combination of complete covariance Gaussian concreteness function. Due to this incorporation, the proposed model tackle most of the limitations of existing classifiers. Some of the well-known features like Mel-frequency cepstral coefficients (MFCC) are extracted in our experiments. The proposed model has been validated and evaluated on two publicly available datasets likes Berlin Database of Emotional Speech (Emo-DB) and the eNTER FACE’05 Audio-Visual Emotion dataset. For validation and comparison against the existing techniques, we utilized 10-fold cross validation scheme. The proposed method achieved significant improvement under the p-value <0.03 for classification. Moreover, we also prove that computational wise, our computation technique is less expensive against state of the art works.

Enhanced blocking block area method for segmentation of continuous speech

Segmentation is one of the important stages in the speech recognition in the type of continuous speech. The segmentation phase serves to break the sentences into words that can be recognized by the computer. The quality of segmentation results can affect the results of the recognition that is done. This research examines the dynamic threshold used in the process of continuous speech segmentation and also proposes an Enhanced Blocking Block Area method in the Indonesian language domain. Three algorithms were compared (K-Means, Fuzzy C-Means, and Otsu) to find the best dynamic threshold and add morphological operations and overlapping to the blocking block area method to obtain the best segmentation accuracy. Based on the results of the research, the Fuzzy C-Means algorithm provides the best threshold results compared to the other two algorithms. By using the Fuzzy C-Means algorithm with the addition of morphology and overlapping, this study can improve the accuracy of continuous speech segmentation in the Indonesian Language from 24% to 90%.

Challenges in ADHD care for ethnic minority children: A review of the current literature.

While attention deficit hyperactivity disorder (ADHD) has been extensively studied in the past decades, the role of social and cultural practices in its assessment, diagnosis, and treatment has been often overlooked. This selective review provides an overview of research that explores social and cultural influences on help-seeking behavior in ethnic minority children with ADHD. Studies were selected that address cultural diversity in three areas of ADHD help-seeking: problem recognition, access to mental health services, and treatment. Special attention was given to studies of treatment selection and adherence in minority groups. Findings suggested that cultural disparities in ADHD care among ethnic minority children occur in the early stages of problem recognition, through service selection, and in the quality of treatment. Ethnic minority children were less likely than their nonminority counterparts to be diagnosed with ADHD and its comorbid conditions and less likely to be prescribed and adhere to stimulant drug treatment. These differences reflect cultural diversity in norms and attitudes towards mental health issues (e.g., fear of social stigma) as well as limited access to qualified health care. Paradoxically, cultural, racial, and language bias may also lead to the overidentification of ethnic minority children as disabled and to higher ratings of ADHD symptoms. This review highlights the importance of sociocultural factors in understanding developmental psychopathology and help-seeking behavior. In addition, it further supports calls for increasing cultural competence in communications during clinical assessment, diagnosis, and treatment in minority communities. Clinical, theoretical, and methodological considerations for future research are discussed.

SOFTWARE FOR RECOGNITION OF CAR NUMBER PLATE

The purpose of this paper is to design and implement an automatic number plate recognition system. The system has still images as the input, and extracts a string corresponding to the plate number, which is used to obtain the output user data from a suitable database. The system extracts data from a license plate and automatically reads it with no prior assumption of background made. License plate extraction is based on plate features, such as texture, and all characters segmented from the plate are passed individually to a character recognition stage for reading. The string output is then used to query a relational database to obtain the desired user data. This particular paper utilizes the intersection of a hat filtered image and a texture mask as the means of locating the number plate within the image. The accuracy of location of the number plate with an image set of 100 images is 68%.

One-shot learning of view-invariant object representations in newborn chicks

Can newborn brains perform one-shot learning? To address this question, we reared newborn chicks in strictly controlled environments containing a single view of a single object, then tested their object recognition performance across 24 uniformly-spaced viewpoints. We found that chicks can build view-invariant object representations from a single view of an object: a case of one-shot learning in newborn brains. Chicks can also build the same view-invariant object representation from different views of an object, showing that newborn brains converge on common object representations from different sets of sensory inputs. Finally, by rearing chicks with larger numbers of object views, we found that chicks develop enhanced recognition for familiar views. These results illuminate the earliest stages of object recognition, revealing (1) powerful one-shot learning that builds invariant object representations from the first views of an object and (2) view-based learning that enriches object representations, producing enhanced recognition for familiar views.

YOLO-face: a real-time face detector

Face detection is one of the important tasks of object detection. Typically detection is the first stage of pattern recognition and identity authentication. In recent years, deep learning-based algorithms in object detection have grown rapidly. These algorithms can be generally divided into two categories, i.e., two-stage detector like Faster R-CNN and one-stage detector like YOLO. Although YOLO and its varieties are not so good as two-stage detectors in terms of accuracy, they outperform the counterparts by a large margin in speed. YOLO performs well when facing normal size objects, but is incapable of detecting small objects. The accuracy decreases notably when dealing with objects that have large-scale changing like faces. Aimed to solve the detection problem of varying face scales, we propose a face detector named YOLO-face based on YOLOv3 to improve the performance for face detection. The present approach includes using anchor boxes more appropriate for face detection and a more precise regression loss function. The improved detector significantly increased accuracy while remaining fast detection speed. Experiments on the WIDER FACE and the FDDB datasets show that our improved algorithm outperforms YOLO and its varieties.

Fast Monocular Visual Place Recognition for Non-Uniform Vehicle Speed and Varying Lighting Environment

This paper presents a novel Fast Monocular Visual Place Recognition (FMPR) with a shallow path-oriented offline learning stage and an online place recognition and tracking stage. FMPR uses a tube of frames with a humanlike key frame recognition to solve place recognition for situations with varying speeds and changing lighting conditions, which are two most commonly encountered situations in real life. We propose an offline learning to analyze the correlation of all video frames in a reference path and to extract effective feature patches of key frames with an offline feature-shifts approach to achieve real-time place recognition. Our recognition results are on the basis of both the instant feature matching of frames and the historical recognition results which impose temporal logic constraints on the movement of a vehicle. Experimental results demonstrate that our proposed method can achieve comparable or even better performance compared with the state-of-the-art methods on different challenging datasets, especially for the case which requires a trade-off between the performance and the processing time. We believe that our FMPR offers a useful alternative to computationally expensive deep learning-based methods especially for applications with battery-powered or resource-limited devices.

Case mixing impedes early lexical access: converging evidence from the masked priming paradigm.

When letters are presented in mixed case (e.g., "PlAnE), word recognition is slowed. This case-mixing effect has been used to argue that early stages of word recognition operate holistically (on the entire visual word form) rather than merely letter-by-letter. Contrary to this holistic view, however, a masked priming study (Perea, Vergara-Martínez, & Gomez, Cognition 142:39-43, 2015) with Spanish words argued that case mixing has no effect on early stages of visual word recognition. Their participants made lexical decisions on an uppercase target (e.g., "PLANE") preceded by an identical prime (e.g., "plane") or an unrelated prime (e.g.,"music"), presented in lowercase or mixed case. Because priming effects (unrelated-identical) were unaffected by case mixing, they concluded that case mixing does not impede early lexical access. We examined whether this finding applies to English words, while also including lowercase targets to prevent a strong bias against holistic word recognition. We found larger priming effects from lowercase primes than mixed-case primes regardless of target case (lowercase vs. uppercase) and whether target case was variedwithin blocks(Experiment 1) or between blocks(Experiment 2). Contrary to Perea et al.'s findings for Spanish, our results suggest an early locus for the case-mixing effect, consistent with theholisticview of word recognition.

Enzymatically functionalized polyaniline thin films produced with one-step electrochemical immobilization and its application in glucose and urea potentiometric biosensors.

Glucose and urea enzymatic biosensors were fabricated. One-step electrochemical immobilization process was used to produce thin polyaniline films with entrapped enzymes. Chronopotentiometric analysis, scanning electron microscopy, electrochemical impedance spectroscopy and optical reflectance spectroscopy were used to determine the structure-property relationship of the functionalized polymeric thin films. The device has a recognition stage connected to a potentiometric field-effect-transistor stage and is based on the measurement of microenvironment pH variation or locally produced ions. Optimization of biosensor fabrication and effective measurement conditions were performed. The optimized films presented sensitivity, linearity and detection range to glucose of 14.6 ± 0.4mV/decade, 99.8% and from 10-4M to 10-1mol/L. Two different biosensors were produced based on the enzymatic reaction of urea with selectivity to ammonium or hydroxyl ions. For ammonium ion selective film, the sensor's parameters were 14.7 ± 0.9mV/decade, 98.2% and from 10-5 to 10-1mol/L. For the hydroxyl ion selective film, the same parameters were 7.4 ± 0.5mV/decade, 98.1% and from 10-5 to 10-1mol/L. The change in the oxidation state of the polymeric matrix explains: i) the large loss of functionality of glucose biosensor in time, ii) the conservation of functionality to the hydroxyl ions for urea biosensor and iii) the selectivity variation of the ammonium ion selective urea biosensor. The results indicate that the polymeric matrix has indeed changeable selectivity, what can be applied in different situations for biosensors production.

Personality Traits and Emotional Word Recognition: An ERP Study.

Recent research has investigated how personality trait differences influence the processing of emotion conveyed by pictures, but limited research has examined the emotion conveyed by words. The present study investigated whether extraversion (extroverts vs. introverts) and neuroticism (high neurotics vs. low neurotics) influence the processing of positive, neutral, and negative words that were matched for arousal. Electroencephalography (EEG) was recorded from healthy participants while they performed a lexical decision task. We found that personality traits influenced emotional word recognition at N400 (300-450 ms) and LPC (450-800 ms). At the earlier (N400) stage, the more extraverted and neurotic a participant was, the more reduced the N400s for the positive words relative to neutral words were. This suggests that the extroverts and high neurotics (i.e., high impulsivity) identified positive content in words during lexical feature retrieval, which facilitated such retrieval. At the later (LPC) stage, both the introverts and high neurotics (i.e., high anxiety) showed greater LPCs to negative than neutral words, indicating their sustained attention and elaborative processing of negative information. These results suggest that extraversion and neuroticism collectively influence different stages of emotional word recognition in a way that is consistent with Gray's biopsychological theory of personality.

Electrome alterations in a plant-pathogen system: Toward early diagnosis

This work aimed to verify the existence of patterns on the electrophysiological systemic responses of tomato plants inoculated with a pathogenic fungus in an environment with controlled light and temperature. Electrical signalling was measured before and after inoculation in the same plants, and data were analysed with time series techniques and approximate multi-scale entropy (ApEn). Machine learning algorithms were utilised in order to classify data before and after infection throughout the five days of experiments. The obtained results have shown that it is possible to distinguish differences in the plant’s electrome activity before and after the fungus inoculation. In some cases, we have found scale invariance quantified by the power law decay in the distribution histogram. We also found a higher degree of internal organisation quantified by ApEn. The results of the classification algorithms achieved higher accuracy of infection detection at the initial stage of pathogen recognition by the plant. Besides, this study showed evidence that long-distance electrical signalling is likely involved in the plant-pathogen interaction, since signals were obtained in the stem and the inoculum applied on the plant leaves. This might be useful for the early detection of plant infections.

Convolutional Long Short-Term Memory Hybrid Networks for Skeletal Based Human Action Recognition

The objective is to develop a time series image representation of the skeletal action data and use it for recognition through a convolutional long short-term deep learning framework. Consequently, Kinect captured human skeletal data is transformed into a Joint Change Distance Image (JCDI) descriptor which maps the time changes in the joints. Subsequently, JCDIs are decoded spatially well with a Convolutional (CNN). Temporal decomposition is executed on long short term memory (LSTM) with data changes along x , y and z position vectors of the skeleton. We propose a combination of CNN and LSTM which maps the spatio temporal information to generate a generalized time series features for recognition. Finally, scores are fused from spatially vibrant CNNs and temporally sound LSTMs for action recognition. Publicly available action datasets such as NTU RGBD, MSR Action, UTKinect and G3D were used as test inputs for experimentation. The results showed a better performance due to spatio temporal modeling at both the representation and the recognition stages when compared to other state-of-the-arts.