Reduce Recognition Errors Research Articles

Research on Intelligent Recognition and Verification System of Football Offside Penalty Based on Artificial Intelligence

An automatic recognition method for ball trajectory and penalty is extensively studied, leveraging computer vision technology. The core hardware components of this system are designed to effectively filter out noise and extract crucial information from the moving track. This enables the system to accurately identify the speed and position of the ball in real-time. To enhance the precision of image segmentation, a novel approach is introduced that utilizes the threshold vector in high-resolution color spaces. This method determines the color of each pixel through bitwise "and" operations between pixel points and subsets. Additionally, a moving window image filtering algorithm is proposed, incorporating trajectory prediction theory. Experimental results demonstrate that this approach significantly reduces recognition errors, resulting in a notably higher accuracy level, making it a promising solution for enhancing the performance of soccer robot systems.

Alignment Knowledge Distillation for Online Streaming Attention-Based Speech Recognition

This article describes an efficient training method for online streaming attention-based encoder-decoder (AED) automatic speech recognition (ASR) systems. AED models have achieved competitive performance in offline scenarios by jointly optimizing all components. They have recently been extended to an online streaming framework via models such as monotonic chunkwise attention (MoChA). However, the elaborate attention calculation process is not robust for long-form speech utterances. Moreover, the sequence-level training objective and time-restricted streaming encoder cause a nonnegligible delay in token emission during inference. To address these problems, we propose CTC synchronous training (CTC-ST), in which CTC alignments are leveraged as a reference for token boundaries to enable a MoChA model to learn optimal monotonic input-output alignments. We formulate a purely end-to-end training objective to synchronize the boundaries of MoChA to those of CTC. The CTC model shares an encoder with the MoChA model to enhance the encoder representation. Moreover, the proposed method provides alignment information learned in the CTC branch to the attention-based decoder. Therefore, CTC-ST can be regarded as self-distillation of alignment knowledge from CTC to MoChA. Experimental evaluations on a variety of benchmark datasets show that the proposed method significantly reduces recognition errors and emission latency simultaneously, especially for long-form and noisy speech. We also compare CTC-ST with several methods that distill alignment knowledge from a hybrid ASR system and show that the CTC-ST can achieve a comparable tradeoff of accuracy and latency without relying on external alignment information. The best MoChA system shows performance comparable to that of RNN-transducer (RNN-T).

Paired Patterns in Logical Analysis of Data for Decision Support in Recognition

Logical analysis of data (LAD), an approach to data analysis based on Boolean functions, combinatorics, and optimization, can be considered one of the methods of interpretable machine learning. A feature of LAD is that, among many patterns, different types of patterns can be identified, for example, prime, strong, spanned, and maximum. This paper proposes a decision-support approach to recognition by sharing different types of patterns to improve the quality of recognition in terms of accuracy, interpretability, and validity. An algorithm was developed to search for pairs of strong patterns (prime and spanned) with the same coverage as the training sample, having the smallest (for the prime pattern) and the largest (for the spanned pattern) number of conditions. The proposed approach leads to a decrease in the number of unrecognized observations (compared with the use of spanned patterns only) by 1.5–2 times (experimental results), to some reduction in recognition errors (compared with the use of prime patterns only) of approximately 1% (depending on the dataset) and makes it possible to assess in more detail the level of confidence of the recognition result due to a refined decision-making scheme that uses the information about the number and type of patterns covering the observation.

SHREC 2022 track on online detection of heterogeneous gestures

This paper presents the outcomes of a contest organized to evaluate methods for the online recognition of heterogeneous gestures from sequences of 3D hand poses. The task is the detection of gestures belonging to a dictionary of 16 classes characterized by different pose and motion features. The dataset features continuous sequences of hand tracking data where the gestures are interleaved with non-significant motions. The data have been captured using the Hololens 2 finger tracking system in a realistic use-case of mixed reality interaction. The evaluation is based not only on the detection performances but also on the latency and the false positives, making it possible to understand the feasibility of practical interaction tools based on the algorithms proposed. The outcomes of the contest’s evaluation demonstrate the necessity of further research to reduce recognition errors, while the computational cost of the algorithms proposed is sufficiently low.

Complex signals alter recognition accuracy and conspecific acceptance thresholds.

Many aspects of behaviour depend on recognition, but accurate recognition is difficult because the traits used for recognition often overlap. For example, brood parasitic birds mimic host eggs, so it is challenging for hosts to discriminate between their own eggs and parasitic eggs. Complex signals that occur in multiple sensory modalities or involve multiple signal components are thought to facilitate accurate recognition. However, we lack models that explore the effect of complex signals on the evolution of recognition systems. Here, we use individual-based models with a genetic algorithm to test how complex signals influence recognition thresholds, signaller phenotypes and receiver responses. The model has three main results. First, complex signals lead to more accurate recognition than simple signals. Second, when two signals provide different amounts of information, receivers will rely on the more informative signal to make recognition decisions and may ignore the less informative signal. As a result, the particular traits used for recognition change over evolutionary time as sender and receiver phenotypes evolve. Third, complex signals are more likely to evolve when recognition errors are high cost than when errors are low cost. Overall, redundant, complex signals are an evolutionarily stable mechanism to reduce recognition errors. This article is part of the theme issue 'Signal detection theory in recognition systems: from evolving models to experimental tests'.

Interventions to Reduce Problems Related to the Readability and Comprehensibility of Drug Packages and Labels: A Systematic Review.

The aim of the study was to evaluate the effectiveness of interventions designed to reduce problems with readability and comprehensibility in drug labels. A systematic review of observational and experimental studies was conducted using MEDLINE (via PubMed), Embase, and Scopus databases. Two independent reviewers selected articles and collected data about study design, interventions, and outcomes. Of 1786 studies screened, 17 were included in this systematic review. The main findings were described according to the type of intervention in the drug package or label: color-coded labels (7 studies); changes in information layout (5 studies); changes in font size (3 studies); and interventions involving look-alike/sound-alike (LASA) drugs (3 studies). Color changes were important both for healthcare professionals to improve safety in hospital practice and for medication users to differentiate between the types of insulin. Layout changes were effective and included highlighting key information, such as medication name, dose, and instructions. Increased font size yielded good results in the recognition and identification of medications mainly for the population with impaired vision. This review was limited by the heterogeneity of interventions and populations studied, as well as the small number of high-quality studies, which prevented meta-analysis across any one intervention or outcome. All studies reported positive results for the proposed interventions. In the case of LASA drug names, the use of resources beyond Tall Man lettering (such as color and highlighting) reduced recognition errors, which may contribute to the development of new strategies to improve safety in the use of LASA drugs.

An inspired haxby brain perceptual model for facial images quality assessment

Face is the most important and most popular biometric used in many identification and verification systems. In these systems, for reducing recognition error rate, the quality of input images need to be as high as possible. Face Image Compliancy verification (FICV) is one of the most essential methods for this purpose. In this research, a brain functionality inspired model is presented for FICV using Haxby model, which is a face visual perception consistent model containing three bilateral areas for three different functionalities. As a result, contribution of this work is presenting a new model, based on human brain functionality, improving the compliancy verification of face images in FICV context. Perceptual understanding of an image is the motivation of most of the quality assessment methods, i.e., the human quality perception is considered as a gold standard and a perfect reference for recognition and quality assessment. The model presented in this work aims to make the operational process of face image quality assessment system closer to the performance of a human expert. Three basic modules have been introduced. Face structural information, for initial information encoding, is simulated by an extended Viola-Jones model. Face image quality assessment is presented by International Civil Aviation Organization (ICAO), in ICAO (ISO / IEC19794 -11) requirements’ compliancy assessment document. Like Haxby model, perception is performed through two distinct functional and neurological pathways, using Hierarchical Maximum pooling (HMAX) and Convolutional Deep Belief Networks (CDBN). Information storing and fetching for training are similar to their corresponding modules in brain. For simulating the brain decision making, the final results of two separate paths are integrated by weighting sum operator. Nine ISO / ICAO requirements were used for testing the model. The simulation results, using AR and PUT databases, shows improvements in six requirements using the proposed method, in comparison with the FICV benchmark.

Bidirectional estimation between context and motion in motion sequence in which context changes

ABSTRACTWe describe a motion recognition method for reducing recognition errors. The method has a two-layer structure: a lower layer for motion recognition that is affected by the distribution of topics used as context information in the upper layer and an upper layer for the topic distribution that is affected by motion recognition in the bottom layer. We introduce an algorithm for the bottom layer to integrate the motion likelihood calculated using a hidden Markov model and motion appearance probabilities obtained by a topic model. We also use a set of particles to estimate and update contexts on the basis of the result of motion recognition in the bottom layer. The set of particles presents a probabilistic distribution of motion topics, and motion recognition and particle update procedures are performed on each particle. We experimented with two types of sequential motion: a sequence of 33 daily motions and complex motion sequences assuming actual observation. The results showed that the proposed method reduced recognition errors and tracked latent topics better than conventional methods did.

Joint dialog act segmentation and recognition in human conversations using attention to dialog context

A dialog act represents the communicative function of an utterance in a conversation, and thus provides informative cues for understanding, managing, and generating dialog. While most spoken dialog systems process user input and system output at the turn level, a single turn can consist of multiple dialog acts in human conversations. Therefore, segmenting turn-level tokens into a meaningful dialog act unit is just as important as recognizing the dialog act. Towards joint segmentation and recognition of dialog acts, we propose an encoder–decoder model featuring joint coding and incorporate contextual information by means of an attentional mechanism. The proposed encoder–decoder outperforms other models in segmentation, and the application of attentions significantly reduces recognition error rates. By combining the encoder–decoder model with contextual attention, we achieve state-of-the-art performance in the joint evaluation of dialog act segmentation and recognition.

An Analytic Survey of End-to-End Speech Recognition Systems

This article presents an analytic survey of various end-to-end speech recognition systems, as well as some approaches to their construction and optimization. We consider models based on connectionist temporal classification (CTC), models based on encoder-decoder architecture with attention mechanism and models using conditional random field (CRF). We also describe integration possibilities with language models at a stage of decoding. We see that such an approach significantly reduces recognition error rates for end-to-end models. A survey of research works in this subject area reveals that end-to-end systems allow achieving results close to that of the state-of-the-art hybrid models. Nevertheless, end-to-end models use simple configuration and demonstrate a high speed of learning and decoding. In addition, we consider popular frameworks and toolkits for creating speech recognition systems.

Noncontact Capacitive Sensing-Based Locomotion Transition Recognition for Amputees With Robotic Transtibial Prostheses.

Recent advancement of robotic transtibial prostheses can restore human ankle dynamics in different terrains. Automatic locomotion transitions of the prosthesis guarantee the amputee's safety and smooth motion. In this paper, we present a noncontact capacitive sensing-based approach for recognizing locomotion transitions of amputees with robotic transtibial prostheses. The proposed sensing system is designed with flexible printed circuit boards which solves the walking instability brought by our previous system when using robotic prosthesis and improves the recognition performance. Six transtibial amputees were recruited and performed tasks of ten locomotion transitions with the robotic prosthesis that we recently constructed. The capacitive sensing system was integrated on the prosthesis and worked in combination with on-prosthesis mechanical sensors. With the cascaded classification method, the proposed system achieved 95.8% average recognition accuracy by support vector machine (SVM) classifier and 94.9% accuracy by quadratic discriminant analysis (QDA) classifier. It could accurately recognize the upcoming locomotion modes from the stance phase of the transition steps. In addition, we proved that adding capacitance signals could significantly reduce recognition errors of the robotic prosthesis in locomotion transition tasks. Our study suggests that the fusion of capacitive sensing system and mechanical sensors is a promising alternative for controlling the robotic transtibial prosthesis.

Perceptual learning eases crowding by reducing recognition errors but not position errors.

When an observer reports a letter flanked by additional letters in the visual periphery, the response errors (the crowding effect) may result from failure to recognize the target letter (recognition errors), from mislocating a correctly recognized target letter at a flanker location (target misplacement errors), or from reporting a flanker as the target letter (flanker substitution errors). Crowding can be reduced through perceptual learning. However, it is not known how perceptual learning operates to reduce crowding. In this study we trained observers with a partial-report task (Experiment 1), in which they reported the central target letter of a three-letter string presented in the visual periphery, or a whole-report task (Experiment 2), in which they reported all three letters in order. We then assessed the impact of training on recognition of both unflanked and flanked targets, with particular attention to how perceptual learning affected the types of errors. Our results show that training improved target recognition but not single-letter recognition, indicating that training indeed affected crowding. However, training did not reduce target misplacement errors or flanker substitution errors. This dissociation between target recognition and flanker substitution errors supports the view that flanker substitution may be more likely a by-product (due to response bias), rather than a cause, of crowding. Moreover, the dissociation is not consistent with hypothesized mechanisms of crowding that would predict reduced positional errors.

On the effectiveness of soft biometrics for increasing face verification rates

The term soft biometrics typically refers to attributes of people such as their gender, the shape of their head or the color of their hair. There is growing interest in soft biometrics as a means of improving automated face recognition since they hold the promise of significantly reducing recognition errors, in part by ruling out illogical choices. This paper concentrates specifically on soft biometrics as opposed to extended attributes, and presents the results from three experiments quantifying performance gains on a difficult face recognition task when standard face recognition algorithms are augmented using soft biometrics. These experiments include (1) a best-case analysis using perfect knowledge of gender and race, (2) support vector machine-based soft biometric classifiers and (3) face shape expressed through an active shape model. All three experiments indicate small improvements may be made when soft biometrics augment an existing algorithm. However, in all cases, the gains were modest. One reason is that false matches are more likely between faces of people sharing the same soft biometric traits. This is to be expected, since face recognition algorithms utilize appearance information, which is the same information used by algorithms designed to assign soft biometric labels to face images.

A Image Retrival Method with Multi-Features Based on Dempster-Shafer Theory

Image content based retrieval is an important research area with application to digital libraries and multimedia databases. Its image characterization and similarity measure must closely follow perceptual characteristics. In this work, a new image retrieval method is proposed by combining color features and texture features based on Dempster-Shafer (D-S) theory. In this proposed method, Multi-color features included color histogram, color moment and color correlogram are used for color analysis, and gabor wavelet features is employed for texture analysis. Then these features are modeled as a mass function in evidence theory, and color and texture detection results are fused at decision-making level. The experimental results show that the proposed method takes advantage of the respective merits of color and texture features and therefore improves retrieval accuracy and reduces recognition error rate.

Detection of speaker individual information using a phoneme effect suppression method

Feature extraction of speaker information from speech signals is a key procedure for exploring individual speaker characteristics and also the most critical part in a speaker recognition system, which needs to preserve individual information while attenuating linguistic information. However, it is difficult to separate individual from linguistic information in a given utterance. For this reason, we investigated a number of potential effects on speaker individual information that arise from differences in articulation due to speaker-specific morphology of the speech organs, comparing English, Chinese and Korean. We found that voiced and unvoiced phonemes have different frequency distributions in speaker information and these effects are consistent across the three languages, while the effect of nasal sounds on speaker individuality is language dependent. Because these differences are confounded with speaker individual information, feature extraction is negatively affected. Accordingly, a new feature extraction method is proposed to more accurately detect speaker individual information by suppressing phoneme-related effects, where the phoneme alignment is required once in constructing a filter bank for phoneme effect suppression, but is not necessary in processing feature extraction. The proposed method was evaluated by implementing it in GMM speaker models for speaker identification experiments. It is shown that the proposed approach outperformed both Mel Frequency Cepstrum Coefficient (MFCC) and the traditional F-ratio (FFCC). The use of the proposed feature has reduced recognition errors by 32.1–67.3% for the three languages compared with MFCC, and by 6.6–31% compared with FFCC. When combining an automatic phoneme aligner with the proposed method, the result demonstrated that the proposed method can detect speaker individuality with about the same accuracy as that based on manual phoneme alignment.

A Pavement Crack Detection Method Combining 2D with 3D Information Based on Dempster‐Shafer Theory

Abstract Pavement cracking is one of the main distresses presented in the road surface. Objective and accurate detection or evaluation for these cracks is an important task in the pavement maintenance and management. In this work, a new pavement crack detection method is proposed by combining two‐dimensional (2D) gray‐scale images and three‐dimensional (3D) laser scanning data based on Dempster‐Shafer (D‐S) theory. In this proposed method, 2D gray‐scale image and 3D laser scanning data are modeled as a mass function in evidence theory, and 2D and 3D detection results for pavement cracks are fused at decision‐making level. The experimental results show that the proposed method takes advantage of the respective merits of 2D images and 3D laser scanning data and therefore improves the pavement crack detection accuracy and reduces recognition error rate compared to 2D image intensity‐based methods.

퍼지 시스템을 이용한 골프 스윙 분류

본 논문에서는 키넥트와 퍼지 시스템을 이용하여 골프 스윙 동작을 7가지 구간으로 분류하는 방법을 제안한다. 퍼지 논리의 입력으로 골프 클럽과 클럽의 헤드 위치를 사용하였으며 이 정보는 키넥트로부터 획득한 골퍼의 관절 정보와 컬러 영상 정보로부터 검출하였다. 제안하는 방법은 크게 신체 관절 추출 모듈, 골프 클럽 검출 및 헤드 추적 모듈, 골프 스윙 동작 분류 모듈로 구성되어 있다. 신체 관절 추출 모듈은 키넥트 센서로부터 검출되는 신체 관절 정보 중 골프 클럽의 검출을 위해 손의 좌표를 추출한다. 두 번째 모듈에서는 손의 좌표를 기준으로 허프 직선 변환 알고리즘을 사용하여 골프 클럽과 골프 클럽의 헤드를 검출한다. 마지막으로 인식 오류를 줄이고 동작별 인식 성능을 향상시키기 위해 퍼지 시스템을 적용하여 골프 스윙 동작을 분류하였다. 실시간 골프 스윙 영상에 대해 제안한 방법의 성능 평가를 시행하였고 제안한 방법은 평균 85.2%의 골프 스윙 동작 분류 신뢰도를 보여줌을 확인하였다. A method to classify a golf swing motion into 7 sections using a Kinect sensor and a fuzzy system is proposed. The inputs to the fuzzy logic are the positions of golf club and its head, which are extracted from the information of golfer's joint position and color information obtained by a Kinect sensor. The proposed method consists of three modules: one for extracting the joint's information, another for detecting and tracking of a golf club, and the other for classifying golf swing motions. The first module extracts the hand's position among the joint information provided by a Kinect sensor. The second module detects the golf club as well as its head with the Hough line transform based on the hand's coordinate. Using a fuzzy logic as a classification engine reduces recognition errors and, consequently, improves the performance of robust classification. From the experiments of real-time video clips, the proposed method shows the reliability of classification by 85.2%.

A paradigm for handwriting-based intelligent tutors

This paper presents the interaction design of, and demonstration of technical feasibility for, intelligent tutoring systems that can accept handwriting input from students. Handwriting and pen input offer several affordances for students that traditional typing-based interactions do not. To illustrate these affordances, we present evidence, from tutoring mathematics, that the ability to enter problem solutions via pen input enables students to record algebraic equations more quickly, more smoothly (fewer errors), and with increased transfer to non-computer-based tasks. Furthermore our evidence shows that students tend to like pen input for these types of problems more than typing. However, a clear downside to introducing handwriting input into intelligent tutors is that the recognition of such input is not reliable. In our work, we have found that handwriting input is more likely to be useful and reliable when context is considered, for example, the context of the problem being solved. We present an intelligent tutoring system for algebra equation solving via pen-based input that is able to use context to decrease recognition errors by 18% and to reduce recognition error recovery interactions to occur on one out of every four problems. We applied user-centered design principles to reduce the negative impact of recognition errors in the following ways: (1) though students handwrite their problem-solving process, they type their final answer to reduce ambiguity for tutoring purposes, and (2) in the small number of cases in which the system must involve the student in recognition error recovery, the interaction focuses on identifying the student’s problem-solving error to keep the emphasis on tutoring. Many potential recognition errors can thus be ignored and distracting interactions are avoided. This work can inform the design of future systems for students using pen and sketch input for math or other topics by motivating the use of context and pragmatics to decrease the impact of recognition errors and put user focus on the task at hand.

A Hybrid Speech Emotion Recognition System Based on Spectral and Prosodic Features

In this paper, we present a hybrid speech emotion recognition system exploiting both spectral and prosodic features in speech. For capturing the emotional information in the spectral domain, we propose a new spectral feature extraction method by applying a novel non-uniform subband processing, instead of the mel-frequency subbands used in Mel-Frequency Cepstral Coefficients (MFCC). For prosodic features, a set of features that are closely correlated with speech emotional states are selected. In the proposed hybrid emotion recognition system, due to the inherently different characteristics of these two kinds of features (e.g., data size), the newly extracted spectral features are modeled by Gaussian Mixture Model (GMM) and the selected prosodic features are modeled by Support Vector Machine (SVM). The final result of the proposed emotion recognition system is obtained by combining the results from these two subsystems. Experimental results show that (1) the proposed non-uniform spectral features are more effective than the traditional MFCC features for emotion recognition; (2) the proposed hybrid emotion recognition system using both spectral and prosodic features yields the relative recognition error reduction rate of 17.0% over the traditional recognition systems using only the spectral features, and 62.3% over those using only the prosodic features.

Exploring Verbal Shadowing as an Effective Method of Enrolling in a Speech Recognition System

The purpose of this research was to investigate the effectiveness of using verbal shadowing as a method of enrolling in a speech recognition system. Verbal shadowing refers to a task whereby a person, while listening to a continuous message, repeats the message aloud word for word. Typically the enrollment process involves users reading an enrollment script aloud and the system processes the speech output to create a personalized acoustic model which increases recognition accuracy. Exploring the feasibility of a non-visually based enrollment technique addresses important human factors issues - accessibility by users with visual impairments or educational limitations. This study found that read enrollments reduced mean recognition errors by 24.70% and shadowed enrollments reduced recognition errors by a mean of 24.64% - a difference of only 0.06%. While this was statistically significantly different, for all practical purposes, shadowing as a method of enrollment proved to be as effective as reading.