• Journal of IKEEE
• /
• v.18 no.4
• /
• pp.485-494
• /
• 2014

Traffic Sign Recognition(TSR) is an important element in an Advanced Driver Assistance System(ADAS). However, many studies related to TSR approaches only in normal daytime environment because a sign's unique color doesn't appear in poor environment such as night time, snow, rain or fog. In this paper, we propose a new TSR algorithm based on machine learning for daytime as well as poor environment. In poor environment, traditional methods which use RGB color region doesn't show good performance. So we extracted sign characteristics using HoG extraction, and detected signs using a Support Vector Machine(SVM). The detected sign is recognized by a decision tree based on 25 reference points in a Normalized RGB system. The detection rate of the proposed system is 96.4% and the recognition rate is 94% when applied in poor environment. The testing was performed on an Intel i5 processor at 3.4 GHz using Full HD resolution images. As a result, the proposed algorithm shows that machine learning based detection and recognition methods can efficiently be used for TSR algorithm even in poor driving environment.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.2
• /
• pp.465-482
• /
• 1997

In this paper, we have studied on cinstructing code-recognition shstem by neural network according to a image process taking the DOT classification code stamped on tire surface.It happened to a few problems that characters distorted in edge by diffused reflection and two adjacent characters take the same label,even very sen- sitive to illumination ofr recognition the stamped them on tire.Thus,this paper would propose the algorithm for tire code under being cinscious of these properties and prove the algorithm drrciency with a simulation.Also,we have suggerted the MCBP network composing of multi-linked recognizers of dffcient identify the DOT code being tire classification code.The MCBP network extracts the projection balue for classifying each character's rdgion after taking out the prjection of each chracter's region on X,Y axis,processes each chracters by taking 7$\times$8 normalization.We have improved error rate 3% through the MCBP network and post-process comparing the DOT code Database. This approach has a accomplished that learming time get's improvenent at 60% and recognition rate has become to 95% from 90% than BckPropagation with including post- processing it has attained greate rates of entire of tire recoggnition at 98%.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.2
• /
• pp.27-34
• /
• 2001

Utterance verification is used in variable vocabulary word recognition to reject the word that does not belong to in-vocabulary word or does not belong to correctly recognized word. Utterance verification is an important technology to design a user-friendly speech recognition system. We propose a new utterance verification algorithm for no-training utterance verification system based on the minimum verification error. First, using PBW (Phonetically Balanced Words) DB (445 words), we create no-training anti-phoneme models which include many PLUs(Phoneme Like Units), so anti-phoneme models have the minimum verification error. Then, for OOV (Out-Of-Vocabulary) rejection, the phoneme-based confidence measure which uses the likelihood between phoneme model (null hypothesis) and anti-phoneme model (alternative hypothesis) is normalized by null hypothesis, so the phoneme-based confidence measure tends to be more robust to OOV rejection. And, the word-based confidence measure which uses the phoneme-based confidence measure has been shown to provide improved detection of near-misses in speech recognition as well as better discrimination between in-vocabularys and OOVs. Using our proposed anti-model and confidence measure, we achieve significant performance improvement; CA (Correctly Accept for In-Vocabulary) is about 89％, and CR (Correctly Reject for OOV) is about 90％, improving about 15-21％ in ERR (Error Reduction Rate).

• Journal of Korea Multimedia Society
• /
• v.11 no.4
• /
• pp.423-433
• /
• 2008

The surface curvatures extracted from the face contain the most important personal facial information. In particular, the face shape using the depth information represents personal features in detail. In this paper, we develop a method for recognizing the range face images by combining the multiple face regions using fuzzy integral. For the proposed approach, the first step tries to find the nose tip that has a protrusion shape on the face from the extracted face area and has to take into consideration of the orientated frontal posture to normalize. Multiple areas are extracted by the depth threshold values from reference point, nose tip. And then, we calculate the curvature features: principal curvature, gaussian curvature, and mean curvature for each region. The second step of approach concerns the application of eigenface and Linear Discriminant Analysis(LDA) method to reduce the dimension and classify. In the last step, the aggregation of the individual classifiers using the fuzzy integral is explained for each region. In the experimental results, using the depth threshold value 40 (DT40) show the highest recognition rate among the regions, and the maximum curvature achieves 98% recognition rate, incase of fuzzy integral.

• Journal of Convergence for Information Technology
• /
• v.11 no.5
• /
• pp.30-37
• /
• 2021

Various devices are connected to the Internet, and attacks using the Internet are occurring. Among such attacks, there are attacks that use malicious URLs to make users access to wrong phishing sites or distribute malicious viruses. Therefore, how to detect such malicious URL attacks is one of the important security issues. Among recent deep learning technologies, neural networks are showing good performance in image recognition, speech recognition, and pattern recognition. This neural network can be applied to research that analyzes and detects patterns of malicious URL characteristics. In this paper, performance analysis according to various parameters was performed on a method of detecting malicious URLs using neural networks. In this paper, malicious URL detection performance was analyzed while changing the activation function, learning rate, and neural network structure. The experimental data was crawled by Alexa top 1 million and Whois to build the data, and the machine learning library used TensorFlow. As a result of the experiment, when the number of layers is 4, the learning rate is 0.005, and the number of nodes in each layer is 100, the accuracy of 97.8% and the f1 score of 92.94% are obtained.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.2
• /
• pp.225-233
• /
• 2024

Active sonar transmits sound waves to detect covertly maneuvering underwater objects and detects the signals reflected back from the target. However, in addition to the target's echo, the active sonar's received signal is mixed with seafloor, sea surface reverberation, biological noise, and other noise, making target recognition difficult. Conventional techniques for detecting signals above a threshold not only cause false detections or miss targets depending on the set threshold, but also have the problem of having to set an appropriate threshold for various underwater environments. To overcome this, research has been conducted on automatic calculation of threshold values through techniques such as Constant False Alarm Rate (CFAR) and application of advanced tracking filters and association techniques, but there are limitations in environments where a significant number of detections occur. As deep learning technology has recently developed, efforts have been made to apply it in the field of underwater target detection, but it is very difficult to acquire active sonar data for discriminator learning, so not only is the data rare, but there are only a very small number of targets and a relatively large number of non-targets. There are difficulties due to the imbalance of data. In this paper, the image of the energy distribution of the detection signal is used, and a classifier is learned in a way that takes into account the imbalance of the data to distinguish between targets and non-targets and added to the existing technique. Through the proposed technique, target misclassification was minimized and non-targets were eliminated, making target recognition easier for active sonar operators. And the effectiveness of the proposed technique was verified through sea experiment data obtained in the East Sea.

• Journal of the Korean Society of Radiology
• /
• v.10 no.6
• /
• pp.443-450
• /
• 2016

In this study, we are using a computer tomography image of the abdomen, as an experimental linear research for the image of the fatty liver patients texture features analysis and computer-aided diagnosis system of implementation using the ROC curve analysis, from the computer tomography image. We tried to provide an objective and reliable diagnostic information of fatty liver to the doctor. Experiments are usually a fatty liver, via the wavelet transform of the abdominal computed tomography images are configured with the experimental image section, shows the results of statistical analysis on six parameters indicating a feature value of the texture. As a result, the entropy, average luminance, strain rate is shown a relatively high recognition rate of 90% or more, the control also, flatness, uniformity showed relatively low recognition rate of about 70%. ROC curve analysis of six parameters are all shown to 0.900 (p = 0.0001) or more, showed meaningful results in the recognition of the disease. Also, to determine the cut-off value for the prediction of disease six parameters. These results are applicable from future abdominal computed tomography images as a preliminary diagnostic article of diseases automatic detection and eventual diagnosis.

• The KIPS Transactions:PartB
• /
• v.9B no.6
• /
• pp.853-862
• /
• 2002

Face Detection can be defined as follows : Given a digitalized arbitrary or image sequence, the goal of face detection is to determine whether or not there is any human face in the image, and if present, return its location, direction, size, and so on. This technique is based on many applications such face recognition facial expression, head gesture and so on, and is one of important qualify factors. But face in an given image is considerably difficult because facial expression, pose, facial size, light conditions and so on change the overall appearance of faces, thereby making it difficult to detect them rapidly and exactly. Therefore, this paper proposes fast and exact face detection which overcomes some restrictions by using neural network. The proposed system can be face detection irrelevant to facial expression, background and pose rapidily. For this. face detection is performed by neural network and detection response time is shortened by reducing search region and decreasing calculation time of neural network. Reduced search region is accomplished by using skin color segment and frame difference. And neural network calculation time is decreased by reducing input vector sire of neural network. Principle Component Analysis (PCA) can reduce the dimension of data. Also, pose estimates in extracted facial image and eye region is located. This result enables to us more informations about face. The experiment measured success rate and process time using the Squared Mahalanobis distance. Both of still images and sequence images was experimented and in case of skin color segment, the result shows different success rate whether or not camera setting. Pose estimation experiments was carried out under same conditions and existence or nonexistence glasses shows different result in eye region detection. The experiment results show satisfactory detection rate and process time for real time system.

• Korean Journal of Human Ecology
• /
• v.10 no.2
• /
• pp.63-75
• /
• 2007

The results of this study on Purchase Patterns and Recognition of Processed Foods of School Meal Service Dietitians of elementary, junior and senior high schools in Chonbuk and Deagu Area are as follows : First, the rate of single cooking of the schools surveyed is 100% in Deagu and 66.9% in Chonbuk, and Chonbuk has more small-meal service schools which caused a higher labor cost than Deagu, Secondly, schools in Deagu has purchased not completely processed vegetables and fish and shells than Chonbuk, and Chonbuk(66.9%) has served more Kim-chi products than Deagu(41.6%). Thirdly, nutritional effects and preference have been considered as school dietitians make plans for the menu. Fourthly, the opinions of the school dietitians about processed food are in the order of high sodium content, convenience and the use of preservative, and Chonbuk has responded positively to the articles of future oriented quality, cooking usage and variety while Deagu has thought of it as an economical. The expected effects from the use of processed foods are in the order of saving labor time and student preferences. Fifthly, meat products have been frequently used and more frozen meat products have been used in Deagu and senior high school than Chonbuk and elementary and junior high schools, last, preferences on processed food are in the order of frozen sea food, noodles. dried sea food, processed vegetable and fruit, dairy goods and others. Senior high school dietitians have preferred packed meat products and other frozen processed food more than elementary and junior high school dietitians. The rate of serving processed food had a different depending on the number of students. In this study, dietitians recognize the harmful effects of processed foods over the merits, which means that the rate of using processed food is low. The proper usage of processed foods is thought to improve the preference of students, to have cooking time shortened and to help manage the meal service sanitarily and efficiently. Therefore, companies producing processed foods should do their best to develop safety and health oriented foods to gain the credibility, and the government should make new regulations for people to purchase and obtain processed foods without any doubt.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.2
• /
• pp.366-373
• /
• 2017

In this paper, we propose a variable weighted average (VWA) learning method in order to improve the performance of the fuzzy ART neural network that has been developed by Grossberg. In a conventional method, the Fast Commit Slow Recode (FCSR), when an input pattern falls in a category, the representative pattern of the category is updated at a fixed learning rate regardless of the degree of similarity of the input pattern. To resolve this issue, a variable learning method proposes reflecting the distance between the input pattern and the representative pattern to reduce the FCSR's category proliferation issue and improve the pattern recognition rate. However, these methods still suffer from the category proliferation issue and limited pattern recognition rate due to inevitable excessive learning created by use of fuzzy AND. The proposed method applies a weighted average learning scheme that reflects the distance between the input pattern and the representative pattern when updating the representative pattern of a category suppressing excessive learning for a representative pattern. Our simulation results show that the newly proposed variable weighted average learning method (VWA) mitigates the category proliferation problem of a fuzzy ART neural network by suppressing excessive learning of a representative pattern in a noisy environment and significantly improves the pattern recognition rates.