• Title/Summary/Keyword: Learning Algorithms

Search Result 2,317, Processing Time 0.03 seconds

Vehicle Headlight and Taillight Recognition in Nighttime using Low-Exposure Camera and Wavelet-based Random Forest (저노출 카메라와 웨이블릿 기반 랜덤 포레스트를 이용한 야간 자동차 전조등 및 후미등 인식)

  • Heo, Duyoung;Kim, Sang Jun;Kwak, Choong Sub;Nam, Jae-Yeal;Ko, Byoung Chul
    • Journal of Broadcast Engineering
    • /
    • v.22 no.3
    • /
    • pp.282-294
    • /
    • 2017
  • In this paper, we propose a novel intelligent headlight control (IHC) system which is durable to various road lights and camera movement caused by vehicle driving. For detecting candidate light blobs, the region of interest (ROI) is decided as front ROI (FROI) and back ROI (BROI) by considering the camera geometry based on perspective range estimation model. Then, light blobs such as headlights, taillights of vehicles, reflection light as well as the surrounding road lighting are segmented using two different adaptive thresholding. From the number of segmented blobs, taillights are first detected using the redness checking and random forest classifier based on Haar-like feature. For the headlight and taillight classification, we use the random forest instead of popular support vector machine or convolutional neural networks for supporting fast learning and testing in real-life applications. Pairing is performed by using the predefined geometric rules, such as vertical coordinate similarity and association check between blobs. The proposed algorithm was successfully applied to various driving sequences in night-time, and the results show that the performance of the proposed algorithms is better than that of recent related works.

Analysis of Disaster Safety Situation Classification Algorithm Based on Natural Language Processing Using 119 Calls Data (119 신고 데이터를 이용한 자연어처리 기반 재난안전 상황 분류 알고리즘 분석)

  • Kwon, Su-Jeong;Kang, Yun-Hee;Lee, Yong-Hak;Lee, Min-Ho;Park, Seung-Ho;Kang, Myung-Ju
    • KIPS Transactions on Software and Data Engineering
    • /
    • v.9 no.10
    • /
    • pp.317-322
    • /
    • 2020
  • Due to the development of artificial intelligence, it is used as a disaster response support system in the field of disaster. Disasters can occur anywhere, anytime. In the event of a disaster, there are four types of reports: fire, rescue, emergency, and other call. Disaster response according to the 119 call also responds differently depending on the type and situation. In this paper, 1280 data set of 119 calls were tested with 3 classes of SVM, NB, k-NN, DT, SGD, and RF situation classification algorithms using a training data set. Classification performance showed the highest performance of 92% and minimum of 77%. In the future, it is necessary to secure an effective data set by disaster in various fields to study disaster response.

Evolution of Neural Network's Structure and Learn Patterns Based on Competitive Co-Evolutionary Method (경쟁적 공진화법에 의한 신경망의 구조와 학습패턴의 진화)

  • Joung, Chi-Sun;Lee, Dong-Wook;Jun, Hyo-Byung;Sim, Kwee-Bo
    • Journal of the Korean Institute of Telematics and Electronics S
    • /
    • v.36S no.1
    • /
    • pp.29-37
    • /
    • 1999
  • In general, the information processing capability of a neural network is determined by its architecture and efficient training patterns. However, there is no systematic method for designing neural network and selecting effective training patterns. Evolutionary Algorithms(EAs) are referred to as the methods of population-based optimization. Therefore, EAs are considered as very efficient methods of optimal system design because they can provide much opportunity for obtaining the global optimal solution. In this paper, we propose a new method for finding the optimal structure of neural networks based on competitive co-evolution, which has two different populations. Each population is called the primary population and the secondary population respectively. The former is composed of the architecture of neural network and the latter is composed of training patterns. These two populations co-evolve competitively each other, that is, the training patterns will evolve to become more difficult for learning of neural networks and the architecture of neural networks will evolve to learn this patterns. This method prevents the system from the limitation of the performance by random design of neural networks and inadequate selection of training patterns. In co-evolutionary method, it is difficult to monitor the progress of co-evolution because the fitness of individuals varies dynamically. So, we also introduce the measurement method. The validity and effectiveness of the proposed method are inspected by applying it to the visual servoing of robot manipulators.

  • PDF

Generation of Efficient Fuzzy Classification Rules Using Evolutionary Algorithm with Data Partition Evaluation (데이터 분할 평가 진화알고리즘을 이용한 효율적인 퍼지 분류규칙의 생성)

  • Ryu, Joung-Woo;Kim, Sung-Eun;Kim, Myung-Won
    • Journal of the Korean Institute of Intelligent Systems
    • /
    • v.18 no.1
    • /
    • pp.32-40
    • /
    • 2008
  • Fuzzy rules are very useful and efficient to describe classification rules especially when the attribute values are continuous and fuzzy in nature. However, it is generally difficult to determine membership functions for generating efficient fuzzy classification rules. In this paper, we propose a method of automatic generation of efficient fuzzy classification rules using evolutionary algorithm. In our method we generate a set of initial membership functions for evolutionary algorithm by supervised clustering the training data set and we evolve the set of initial membership functions in order to generate fuzzy classification rules taking into consideration both classification accuracy and rule comprehensibility. To reduce time to evaluate an individual we also propose an evolutionary algorithm with data partition evaluation in which the training data set is partitioned into a number of subsets and individuals are evaluated using a randomly selected subset of data at a time instead of the whole training data set. We experimented our algorithm with the UCI learning data sets, the experiment results showed that our method was more efficient at average compared with the existing algorithms. For the evolutionary algorithm with data partition evaluation, we experimented with our method over the intrusion detection data of KDD'99 Cup, and confirmed that evaluation time was reduced by about 70%. Compared with the KDD'99 Cup winner, the accuracy was increased by 1.54% while the cost was reduced by 20.8%.

Adaptation of Neural Network based Intelligent Characters to Change of Game Environments (신경망 지능 캐릭터의 게임 환경 변화에 대한 적응 방법)

  • Cho Byeong-heon;Jung Sung-hoon;Sung Yeong-rak;Oh Ha-ryoung
    • Journal of the Institute of Electronics Engineers of Korea CI
    • /
    • v.42 no.3 s.303
    • /
    • pp.17-28
    • /
    • 2005
  • Recently intelligent characters in computer games have been an important element more and more because they continually stimulate gamers' interests. As a typical method for implementing such intelligent characters, neural networks have been used for training action patterns of opponent's characters and game rules. However, some of the game rules can be abruptly changed and action properties of garners in on-line game environments are quite different according to gamers. In this paper, we address how a neural network adapts to those environmental changes. Our adaptation solution includes two components: an individual adaptation mechanism and a group adaptation mechanism. With the individual adaptation algorithm, an intelligent character steadily checks its game score, assesses the environmental change with taking into consideration of the lastly earned scores, and initiates a new learning process when a change is detected. In multi-user games, including massively multiple on-line games, intelligent characters confront diverse opponents that have various action patterns and strategies depending on the gamers controlling the opponents. The group adaptation algorithm controls the birth of intelligent characters to conserve an equilibrium state of a game world by using a genetic algorithm. To show the performance of the proposed schemes, we implement a simple fighting action game and experiment on it with changing game rules and opponent characters' action patterns. The experimental results show that the proposed algorithms are able to make intelligent characters adapt themselves to the change.

Classification Modeling for Predicting Medical Subjects using Patients' Subjective Symptom Text (환자의 주관적 증상 텍스트에 대한 진료과목 분류 모델 구축)

  • Lee, Seohee;Kang, Juyoung
    • The Journal of Bigdata
    • /
    • v.6 no.1
    • /
    • pp.51-62
    • /
    • 2021
  • In the field of medical artificial intelligence, there have been a lot of researches on disease prediction and classification algorithms that can help doctors judge, but relatively less interested in artificial intelligence that can help medical consumers acquire and judge information. The fact that more than 150,000 questions have been asked about which hospital to go over the past year in NAVER portal will be a testament to the need to provide medical information suitable for medical consumers. Therefore, in this study, we wanted to establish a classification model that classifies 8 medical subjects for symptom text directly described by patients which was collected from NAVER portal to help consumers choose appropriate medical subjects for their symptoms. In order to ensure the validity of the data involving patients' subject matter, we conducted similarity measurements between objective symptom text (typical symptoms by medical subjects organized by the Seoul Emergency Medical Information Center) and subjective symptoms (NAVER data). Similarity measurements demonstrated that if the two texts were symptoms of the same medical subject, they had relatively higher similarity than symptomatic texts from different medical subjects. Following the above procedure, the classification model was constructed using a ridge regression model for subjective symptom text that obtained validity, resulting in an accuracy of 0.73.

A Study on the Forecasting Trend of Apartment Prices: Focusing on Government Policy, Economy, Supply and Demand Characteristics (아파트 매매가 추이 예측에 관한 연구: 정부 정책, 경제, 수요·공급 속성을 중심으로)

  • Lee, Jung-Mok;Choi, Su An;Yu, Su-Han;Kim, Seonghun;Kim, Tae-Jun;Yu, Jong-Pil
    • The Journal of Bigdata
    • /
    • v.6 no.1
    • /
    • pp.91-113
    • /
    • 2021
  • Despite the influence of real estate in the Korean asset market, it is not easy to predict market trends, and among them, apartments are not easy to predict because they are both residential spaces and contain investment properties. Factors affecting apartment prices vary and regional characteristics should also be considered. This study was conducted to compare the factors and characteristics that affect apartment prices in Seoul as a whole, 3 Gangnam districts, Nowon, Dobong, Gangbuk, Geumcheon, Gwanak and Guro districts and to understand the possibility of price prediction based on this. The analysis used machine learning algorithms such as neural networks, CHAID, linear regression, and random forests. The most important factor affecting the average selling price of all apartments in Seoul was the government's policy element, and easing policies such as easing transaction regulations and easing financial regulations were highly influential. In the case of the three Gangnam districts, the policy influence was low, and in the case of Gangnam-gu District, housing supply was the most important factor. On the other hand, 6 mid-lower-level districts saw government policies act as important variables and were commonly influenced by financial regulatory policies.

Variation of Seasonal Groundwater Recharge Analyzed Using Landsat-8 OLI Data and a CART Algorithm (CART알고리즘과 Landsat-8 위성영상 분석을 통한 계절별 지하수함양량 변화)

  • Park, Seunghyuk;Jeong, Gyo-Cheol
    • The Journal of Engineering Geology
    • /
    • v.31 no.3
    • /
    • pp.395-432
    • /
    • 2021
  • Groundwater recharge rates vary widely by location and with time. They are difficult to measure directly and are thus often estimated using simulations. This study employed frequency and regression analysis and a classification and regression tree (CART) algorithm in a machine learning method to estimate groundwater recharge. CART algorithms are considered for the distribution of precipitation by subbasin (PCP), geomorphological data, indices of the relationship between vegetation and landuse, and soil type. The considered geomorphological data were digital elevaion model (DEM), surface slope (SLOP), surface aspect (ASPT), and indices were the perpendicular vegetation index (PVI), normalized difference vegetation index (NDVI), normalized difference tillage index (NDTI), normalized difference residue index (NDRI). The spatio-temperal distribution of groundwater recharge in the SWAT-MOD-FLOW program, was classified as group 4, run in R, sampled for random and a model trained its groundwater recharge was predicted by CART condidering modified PVI, NDVI, NDTI, NDRI, PCP, and geomorphological data. To assess inter-rater reliability for group 4 groundwater recharge, the Kappa coefficient and overall accuracy and confusion matrix using K-fold cross-validation were calculated. The model obtained a Kappa coefficient of 0.3-0.6 and an overall accuracy of 0.5-0.7, indicating that the proposed model for estimating groundwater recharge with respect to soil type and vegetation cover is quite reliable.

Performance Analysis of Object Detection Neural Network According to Compression Ratio of RGB and IR Images (RGB와 IR 영상의 압축률에 따른 객체 탐지 신경망 성능 분석)

  • Lee, Yegi;Kim, Shin;Lim, Hanshin;Lee, Hee Kyung;Choo, Hyon-Gon;Seo, Jeongil;Yoon, Kyoungro
    • Journal of Broadcast Engineering
    • /
    • v.26 no.2
    • /
    • pp.155-166
    • /
    • 2021
  • Most object detection algorithms are studied based on RGB images. Because the RGB cameras are capturing images based on light, however, the object detection performance is poor when the light condition is not good, e.g., at night or foggy days. On the other hand, high-quality infrared(IR) images regardless of weather condition and light can be acquired because IR images are captured by an IR sensor that makes images with heat information. In this paper, we performed the object detection algorithm based on the compression ratio in RGB and IR images to show the detection capabilities. We selected RGB and IR images that were taken at night from the Free FLIR Thermal dataset for the ADAS(Advanced Driver Assistance Systems) research. We used the pre-trained object detection network for RGB images and a fine-tuned network that is tuned based on night RGB and IR images. Experimental results show that higher object detection performance can be acquired using IR images than using RGB images in both networks.

Vizrt Engine-Based Virtual Reality Graphics Algorithm A Study on the Basic Practical Training Method (Vizrt 엔진 기반 가상현실 그래픽 알고리즘과 기초 실습 교육 방식의 연구)

  • Cho, Hyun Kyung
    • The Journal of the Convergence on Culture Technology
    • /
    • v.5 no.3
    • /
    • pp.197-202
    • /
    • 2019
  • In the era of the fourth revolution, interest in content production using proven engines in the broadcasting sector, such as Vizrt, is growing. The new visual effects required in the 5G era are critical to content production training. Vizrt has a good production time utility and affordability for broadcast and media content. In this paper, we are going to use this to present a practical case of the theorem and application of the basic training course in the production of virtual content, and to present the basic training direction. In the introduction, the graphic algorithm analyzed and studied the characteristics and environmental factors of the Vizrt engine. In this paper, the production process was studied separately, and the work carried out through engine implementation was presented. The VS Studio Foundation was provided as a practical production case at each stage. The Vizrt engine operator process is important in graphic approach and application, and through the results of the lecture, the method of understanding and implementing algorithms for virtual reality perspective suitable for basic learning was studied. Based on practice, the research method of main theory was to create Vizrt contents specialized in 5G contents work in each sector and to implement graphic production in new areas from contents image. Through this study, we came to the conclusion of the basic training method through virtual reality content work based on Vizrt by practicing content creation according to the subject. It also proposes the effect of creating Vizrt content and the direction of building Vizrt basic training courses.