• 스마트미디어저널
• /
• 제11권8호
• /
• pp.84-92
• /
• 2022

Precision livestock monitoring promises greater management efficiency for farmers and higher welfare standards for animals. Recent studies on video-based animal activity recognition and tracking have shown promising solutions for understanding animal behavior. To achieve that, surveillance cameras are installed diagonally above the barn in a typical cattle farm setup to monitor animals constantly. Under these circumstances, tracking individuals requires addressing challenges such as occlusion and visual appearance, which are the main reasons for track breakage and increased misidentification of animals. This paper presents a framework for multi-cattle tracking in closed barns with appearance and motion models. To overcome the above challenges, we modify the DeepSORT algorithm to achieve higher tracking accuracy by three contributions. First, we reduce the weight of appearance information. Second, we use an Ensemble Kalman Filter to predict the random motion information of cattle. Third, we propose a supplementary matching algorithm that compares the absolute cattle position in the barn to reassign lost tracks. The main idea of the matching algorithm assumes that the number of cattle is fixed in the barn, so the edge of the barn is where new trajectories are most likely to emerge. Experimental results are performed on our dataset collected on two cattle farms. Our algorithm achieves 70.37%, 77.39%, and 81.74% performance on HOTA, AssA, and IDF1, representing an improvement of 1.53%, 4.17%, and 0.96%, respectively, compared to the original method.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.141-151
• /
• 2022

In recent years, the industry and research communities have focused on developing autonomous crack inspection approaches, which mainly include image acquisition and crack detection. In these approaches, mobile devices such as cameras, drones or smartphones are utilized as sensing platforms to acquire structural images, and the deep learning (DL)-based methods are being developed as important crack detection approaches. However, the process of image acquisition and collection is time-consuming, which delays the inspection. Also, the present mobile devices such as smartphones can be not only a sensing platform but also a computing platform that can be embedded with deep neural networks (DNNs) to conduct on-site crack detection. Due to the limited computing resources of mobile devices, the size of the DNNs should be reduced to improve the computational efficiency. In this study, an architecture called pruned crack recognition network (PCR-Net) was developed for the detection of structural cracks. A dataset containing 11000 images was established based on the raw images from bridge inspections. A pruning method was introduced to reduce the size of the base architecture for the optimization of the model size. Comparative studies were conducted with image processing techniques (IPTs) and other DNNs for the evaluation of the performance of the proposed PCR-Net. Furthermore, a modularly designed framework that integrated the PCR-Net was developed to realize a DL-based crack detection application for smartphones. Finally, on-site crack detection experiments were carried out to validate the performance of the developed system of smartphone-based detection of structural cracks.

• 한국항행학회논문지
• /
• 제21권4호
• /
• pp.365-370
• /
• 2017

본 연구는 전력수요 패턴이 다른 평일과 특수일 데이터가 가지는 상관관계를 분석하여, 별도의 데이터 셋을 구축하고, 각 데이터 셋에 적합한 딥 러닝 네트워크를 이용하여, 전력수요예측 오차를 감소하는 방안을 제시하였다. 또한, 기본적인 전력수요 예측요소인 기상요소에 환경요소, 구분요소 등 다양한 예측요소를 추가하여 예측율을 향상하는 방안을 제시하였다. 전체데이터는 시계열 데이터 학습에 적합한 LSTM을 이용하여 전력수요예측을 하였으며, 특수일 데이터는 DNN을 이용하여 전력수요예측을 하였다. 실험결과 기상요소 이외의 예측요소 추가를 통해 예측율이 향상되었다. 전체 데이터 셋의 평균 RMSE는 LSTM이 0.2597이며, DNN이 0.5474로 LSTM이 우수한 예측율을 보였다. 특수일 데이터 셋의 평균 RMSE는 0.2201로 DNN이 LSTM보다 우수한 예측율을 보였다. 또한, 전체 데이터 셋의 LSTM의 MAPE는 2.74 %이며, 특수 일의 MAPE는 3.07 %를 나타냈다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2021년도 추계학술대회
• /
• pp.218-220
• /
• 2021

본 논문에서는 CAR(content adaptive resampler)로 축소된 저해상도 이미지를 직접 다른 모델에 여러가지 방식으로 훈련을 시켜 성능을 개선시키고자 하였다. 본 논문에서는 단일 영상 super resolution 에 관하여 여러 기술이 존재하는 상황에 더 나은 기술을 테스트하려 하고 그를 위해 과거의 모델들에 대한 이해가 필요하여 이를 구현하였다. 현재 가장 뛰어난 성능을 보이고 있는 모델 중의 하나인 CAR 에서 복원 전 이미지를 사용하여 훈련을 시키면 더 나은 성능의 모델을 만들 수 있을 것이라고 가정하고 다양한 훈련을 통해 성능을 개선시키고자 하였다.

• 한국컴퓨터정보학회논문지
• /
• 제26권10호
• /
• pp.19-26
• /
• 2021

딥러닝 기반 악성코드 탐지 및 분류모델의 성능은 특성집합을 어떻게 구성하느냐에 따라 크게 좌우된다. 본 논문에서는 CNN 기반의 안드로이드 악성코드 탐지 시 탐지성능을 극대화할 수 있는 최적의 특성집합(feature set)을 선정하는 방법을 제안한다. 특성집합에 포함될 특성은 기계학습 및 딥러닝에서 특성추출을 위해 널리 사용되는 Chi-Square test 알고리즘을 사용하여 선정하였다. CICANDMAL2017 데이터세트를 대상으로 선정된 36개의 특성을 이용하여 CNN 모델을 학습시킨 후 악성코드 탐지성능을 측정한 결과 이진분류에서는 99.99%, 다중분류에서는 98.55%의 Accuracy를 달성하였다.

• International Journal of Computer Science & Network Security
• /
• 제21권12호
• /
• pp.213-218
• /
• 2021

This paper proposes a technique for detecting a significant threat that attempts to get sensitive and confidential information such as usernames, passwords, credit card information, and more to target an individual or organization. By definition, a phishing attack happens when malicious people pose as trusted entities to fraudulently obtain user data. Phishing is classified as a type of social engineering attack. For a phishing attack to happen, a victim must be convinced to open an email or a direct message [1]. The email or direct message will contain a link that the victim will be required to click on. The aim of the attack is usually to install malicious software or to freeze a system. In other instances, the attackers will threaten to reveal sensitive information obtained from the victim. Phishing attacks can have devastating effects on the victim. Sensitive and confidential information can find its way into the hands of malicious people. Another devastating effect of phishing attacks is identity theft [1]. Attackers may impersonate the victim to make unauthorized purchases. Victims also complain of loss of funds when attackers access their credit card information. The proposed method has two major subsystems: (1) Data collection: different websites have been collected as a big data corresponding to normal and phishing dataset, and (2) distributed detection system: different artificial algorithms are used: a neural network algorithm and machine learning. The Amazon cloud was used for running the cluster with different cores of machines. The experiment results of the proposed system achieved very good accuracy and detection rate as well.

• 스마트미디어저널
• /
• 제8권2호
• /
• pp.46-57
• /
• 2019

In recent research, deep learning-based methods have achieved state-of-the-art performance in various computer vision tasks. However, these methods are commonly supervised, and require huge amounts of annotated data to train. Acquisition of data demands an additional costly effort, particularly for the tasks where it becomes challenging to obtain large amounts of data considering the time constraints and the requirement of professional human diligence. In this paper, we present a data level synthetic sampling solution to learn from small and imbalanced data sets using Generative Adversarial Networks (GANs). The reason for using GANs are the challenges posed in various fields to manage with the small datasets and fluctuating amounts of samples per class. As a result, we present an approach that can improve learning with respect to data distributions, reducing the partiality introduced by class imbalance and hence shifting the classification decision boundary towards more accurate results. Our novel method is demonstrated on a small dataset of 2789 tomato plant disease images, highly corrupted with class imbalance in 9 disease categories. Moreover, we evaluate our results in terms of different metrics and compare the quality of these results for distinct classes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권6호
• /
• pp.2018-2043
• /
• 2022

Nowadays, COVID-19 infections are influencing our daily lives which have spread globally. The major symptoms' of COVID-19 are dry cough, sore throat, and fever which in turn to critical complications like multi organs failure, acute respiratory distress syndrome, etc. Therefore, to hinder the spread of COVID-19, a Computerized Doughty Predictor Framework (CDPF) is developed to yield benefits in monitoring the progression of disease from Chest CT images which will reduce the mortality rates significantly. The proposed framework CDPF employs Convolutional Neural Network (CNN) as a feature extractor to extract the features from CT images. Subsequently, the extracted features are fed into the Adaptive Dragonfly Algorithm (ADA) to extract the most significant features which will smoothly drive the diagnosing of the COVID and Non-COVID cases with the support of Doughty Learners (DL). This paper uses the publicly available SARS-CoV-2 and Github COVID CT dataset which contains 2482 and 812 CT images with two class labels COVID⁺ and COVI^-. The performance of CDPF is evaluated against existing state of art approaches, which shows the superiority of CDPF with the diagnosis accuracy of about 99.76%.

• 로봇학회논문지
• /
• 제18권2호
• /
• pp.143-154
• /
• 2023

In this study, a fast motion planning method for the swing motion of a 6x6 wheel-legged robot to traverse large obstacles and gaps is proposed. The motion planning method presented in the previous paper, which was based on trajectory optimization, took up to tens of seconds and was limited to two-dimensional, structured vertical obstacles and trenches. A deep neural network based on one-dimensional Convolutional Neural Network (CNN) is introduced to generate keyframes, which are then used to represent smooth reference commands for the six leg angles along the robot's path. The network is initially trained using the behavioral cloning method with a dataset gathered from previous simulation results of the trajectory optimization. Its performance is then improved through reinforcement learning, using a one-step REINFORCE algorithm. The trained model has increased the speed of motion planning by up to 820 times and improved the success rates of obstacle crossing under harsh conditions, such as low friction and high roughness.

• 한국정보통신학회논문지
• /
• 제25권11호
• /
• pp.1505-1511
• /
• 2021

자연어 처리 분야 내 다양한 작업들에서 높은 성능을 보인 사전 학습된 모델은 대량의 말뭉치를 이용하여 문장들의 언어학적 패턴을 스스로 학습함으로써 입력 문장 내 각 토큰들을 적절한 특징 벡터로 표현할 수 있다는 장점을 갖고 있다. 이러한 사전 학습된 모델의 학습에 필요한 말뭉치를 구축하는 방법 중 웹 크롤러를 이용하여 수집한 경우 웹사이트에 존재하는 문장은 다양한 패턴을 갖고 있기 때문에 문장의 일부 또는 전체에 불필요한 단어가 포함되어 있을 수 있다. 본 논문에서는 웹으로부터 수집한 말뭉치에 대해 신경망 모델을 이용하여 불필요한 단어가 포함된 문장을 필터링하기 위한 데이터 셋 구축 방법에 대해 제안한다. 그 결과, 총 2,330개의 문장을 포함한 데이터 셋을 구축하였다. 또한 신경망 모델을 이용하여 구축한 데이터 셋을 학습시켜 성능을 평가하였으며, BERT 모델이 평가 데이터에 대해 93.75%의 정확도로 가장 높은 성능을 보였다.