Crowd behavior simulations have been studied to further accelerated and refined by parallelism by inducing agent-interacting forces into the image field representing the forces of attraction and repulsion. However, it was difficult to consider rapidly changing environments such as fire situations in buildings because texture images must be generated in advance simulation starts and simulations can only be performed in 2D spaces. In this paper, we propose a crowd agent behavior simulation method based on agent's 3D affordance field for flexible agent behavior in variable geomorphological environments in 3D space. The proposed method generates 3D affordance field related to agents and sensors in 3D space and defines the agent behavior in 3D space for the crowd behavior simulation based on an image-inducing field to a 3D space. Experimental results verified that our method enables the development of large-scale crowd behavior simulations that are flexible to various fire evacuation situations in 3D virtual spaces.

Strabismus is one of the most common disease that might be associated with vision impairment. Especially in infants and children, it is critical to detect strabismus at an early age because uncorrected strabismus may go on to develop amblyopia. To this end, ophthalmologists usually perform the Hirschberg test, which observes corneal light reflex (CLR) to determine the presence and type of strabismus. However, this test is usually done manually in a hospital, which might be difficult for patients who live in a remote area with poor medical access. To address this issue, we propose an automatic strabismus screening method that calculates the CLR ratio to determine the presence of strabismus based on image processing. In particular, the method first employs a pre-trained face detection model and a 68 facial landmarks detector to extract the eye region image. The data points located in the limbus are then collected, and the least square method is applied to obtain the center coordinates of the iris. Finally, the coordinate of the reflective light point center within the iris is extracted and used to calculate the CLR ratio with the coordinate of iris edges. Experimental results with several images demonstrate that the proposed method can be a promising solution to provide strabismus screening for patients who cannot visit hospitals.

In this paper, we propose methods for character recognition and classification for table items through deep learning. First, table areas are detected in a document image through CNN. After that, table areas are separated by separators such as vertical lines. The text in document is recognized through a neural network combined with CNN and RNN. To correct errors in the character recognition, multiple candidates for the recognized result are provided for a sentence which has low recognition accuracy.

With the continuous development of deep learning, human behavior recognition algorithms have achieved good results. However, in a multi-person recognition environment, the complex behavior environment poses a great challenge to the efficiency of recognition. To this end, this paper proposes a multi-person pose estimation model. First of all, the human detectors in the top-down framework mostly use the two-stage target detection model, which runs slow down. The single-stage YOLOv3 target detection model is used to effectively improve the running speed and the generalization of the model. Depth separable convolution, which further improves the speed of target detection and improves the model's ability to extract target proposed regions; Secondly, based on the feature pyramid network combined with context semantic information in the pose estimation model, the OHEM algorithm is used to solve difficult key point detection problems, and the accuracy of multi-person pose estimation is improved; Finally, the Euclidean distance is used to calculate the spatial distance between key points, to determine the similarity of postures in the frame, and to eliminate redundant postures.

The treatments for symptoms of Alzheimer's disease are being provided and for the early diagnosis several researches are undergoing. In this regard, by using T1-weighted images several classification techniques had been proposed to distinguish among AD, MCI, and Healthy Control (HC) patients. In this paper, we also used some traditional Machine Learning (ML) approaches in order to diagnose the AD. This paper consists of an improvised feature selection method which is used to reduce the model complexity which accounted an issue while utilizing the ML approaches. In our presented work, combination of subcortical and cortical features of 308 subjects of ADNI dataset has been used to diagnose AD using structural magnetic resonance (sMRI) images. Three classification experiments were performed: binary classification. i.e., AD vs eMCI, AD vs lMCI, and AD vs HC. Proposed Feature Selection method consist of a combination of Principal Component Analysis and Recursive Feature Elimination method that has been used to reduce the dimension size and selection of best features simultaneously. Experiment on the dataset demonstrated that SVM is best suited for the AD vs lMCI, AD vs HC, and AD vs eMCI classification with the accuracy of 95.83%, 97.83%, and 97.87% respectively.

Wi-Fi Round-trip timing (RTT) based location estimation technology estimates the distance between the user and the AP based on the transmission and reception time of the signal. This is because reception instability and signal distortion are greater than that of a Received Signal Strength Indicator (RSSI) based fingerprint in an indoor NLOS environment, resulting in a large position error due to multipath fading. To solve this problem, in this paper, we propose an end-to-end based WiFi Trilateration Net (WTN) that combines neural network-based RTT correction and trilateral positioning network, respectively. The proposed WTN is composed of an RNN-based correction network to improve the RTT distance accuracy and a neural network-based trilateral positioning network for real-time positioning implemented in an end-to-end structure. The proposed network improves learning efficiency by changing the trilateral positioning algorithm, which cannot be learned through differentiation due to mathematical operations, to a neural network. In addition, in order to increase the stability of the TOA based RTT, a correction network is applied in the scanning step to collect reliable distance estimation values from each RTT AP.

Resting-state Functional Magnetic Resonance Imaging(fMRI) data detects the temporal correlations in Blood Oxygen Level Dependent(BOLD) signal and these temporal correlations are regarded to reflect intrinsic cortical connectivity, which is deactivated during attention demanding, non-self referential tasks, called Default Mode Network(DMN). The relationship between fMRI and anatomical connectivity has not been studied in detail, however, the preceded studies have tried to clarify this relationship using Diffusion Tensor Imaging(DTI) and fMRI. These studies use method that fMRI data assists DTI data or vice versa and it is used as guider to perform DTI tractography on the brain image. In this study, we hypothesized that functional connectivity in resting state would reflect anatomical connectivity of DMN and the combined images include information of fMRI and DTI showed visible connection between brain regions related in DMN. In the previous study, functional connectivity was determined by subjective region of interest method. However, in this study, functional connectivity was determined by objective and advanced method through Independent Component Analysis. There was a stronger connection between Posterior Congulate Cortex(PCC) and PHG(Parahippocampa Gyrus) than Anterior Cingulate Cortex(ACC) and PCC. This technique might be used in several clinical field and will be the basis for future studies related to aging and the brain diseases, which are needed to be translated not only functional connectivity, but structural connectivity.

The Faster RCNN-based shellfish recognition algorithm is introduced for shellfish recognition studies that currently do not have any deep learning-based algorithms in a practical setting. The original feature extraction module is replaced by DenseNet, which fuses multi-level feature data and optimises the NMS algorithm, network depth and merging method; overcoming the omission of shellfish overlap, multiple shellfish and insufficient light, effectively solving the problem of low shellfish classification accuracy. In the complexifier test environment, the test accuracy was improved by nearly 4%. Higher testing accuracy was achieved compared to the original testing algorithm. This provides favourable technical support for future applications of the improved Faster RCNN approach to seafood quality classification.

As one of the most widely used technology in the world right now, Face recognition has already received widespread attention by all the researcher and institutes. It has been used in many fields such as safety protection, surveillance system, crime control and even in our ordinary life such as home security and so on. This technology with today's technology has advantages such as high connectivity and real time transformation. But we still need to improve its recognition rate, reaction time and also reduce impact of different environmental status to the whole system. So in this paper we proposed a face recognition system model with improved CNN which combining the characteristics of flat network and residual network, integrated learning, simplify network structure and enhance portability and also improve the recognition accuracy. We also used AR and ORL database to do the experiment and result shows higher recognition rate, efficiency and robustness for different image conditions.

The current trend of entertainment programs is an "Sub-character." This study is to find out which strategies were chosen, considering the entertainment program "Hang Out With Yoo" as the beginning and front-runner of craze for Sub-characters. The following results were deduced: implementation of trendy characters, aggressive acceptance of growth narrative and expansion of episodes. The success or failure of an entertainment program depends on the implementation method of differentiated characters. At this point in time when Sub-characters are continuously developed, this program is meaningful in that it has presented the strategies for Sub-characters reflecting the demand of audience.

The zombies in AVG games are not divided into different game occupations. To address this issue, this paper starts by selecting those popular zombie games as the research object. Then, a number of samples will be collected from the roles of zombie and these different roles of zombie will be classified by profession depending on the characteristics of these roles. On this basis, the roles will be given a new title to describe their profession. Then, semantic analysis will be conducted to search for the words that can be used to describe the roles of zombie. Besides, there will be 10 experts in design and 10 professional zombie game players invited to choose from the collected adjectives. The most selected words will be used in this study to conduct cognition assessment, before the making of questionnaire. Then, the questionnaire data will be collected and converted into a statistical model form for indicating the value of ability by group and profession.

The purpose of this study is to evaluate visual arts in interactive technology in art museum applications and to analyze the impact relationship between interactive technology and the five senses. This study was conducted to survey respondents who had used art museum applications. The results are as follows. First, this study evaluates the differences in perception of art museum applications according to the general characteristics of the respondents. They show that there are differences in gender, marital status, age, and social income. Second, this study identifies the impact of five senses and synesthesia on interaction design among emotional design elements. They reveal that visual, auditory, tactile components, and synesthesia have significant effects on interactive design. This study reveals that emotional design elements of art museum applications affect interaction design. Also, it suggests that research on interaction design reflecting five senses is continuously needed to improve audience satisfaction and revitalize art museum applications.

Due to COVID-19, the non-face-to-face content market is growing rapidly. However, most of the non-face-to-face content such as webtoons and web novels are produced based on the traditional culture of other countries, not Korean traditional culture. The biggest cause of this situation is the lack of reference materials for creating based on Korean traditional culture. Therefore, the need for materials on traditional Korean culture that can be used for content creation is emerging. In this paper, we propose a generation model of source data based on traditional folk paintings through the fusion of traditional Korean folk paintings and AI technology. The proposed model secures basic data based on folk tales, analyzes the style and characteristics of folk tales, and converts historical backgrounds and various stories related to folk tales into data. In addition, using the built data, various new stories are created based on AI technology. The proposed model is highly utilized in that it provides a foundation for new creation based on Korean traditional folk painting and AI technology.