• Journal of Broadcast Engineering
• /
• v.21 no.4
• /
• pp.609-620
• /
• 2016

Recently, there are active studies on a forward collision warning system to prevent the accidents and improve convenience of drivers. For collision evasion, the vehicle detection system is required. In general, existing learning-based vehicle detection methods use the entire appearance of the vehicles from rear-view images, so that each vehicle types should be learned separately since they have distinct rear-view appearance regarding the types. To overcome such shortcoming, we learn Haar-like features from the lower part of the vehicles which contain tail lights to detect vehicles leveraging the fact that the lower part is consistent regardless of vehicle types. As a verification procedure, we detect tail lights to distinguish actual vehicles and non-vehicles. If candidates are too small to detect the tail lights, we use HOG(Histogram Of Gradient) feature and SVM(Support Vector Machine) classifier to reduce false alarms. The proposed forward vehicle detection method shows accuracy of 95% even in the complicated images with many buildings by the road, regardless of vehicle types.

• Korean Journal of Remote Sensing
• /
• v.15 no.1
• /
• pp.39-50
• /
• 1999

It is complicate and time-consuming process to classify a multi-band satellite imagery according to the application. In addition, classification rate sensitively depends on the selection of training data set and features in a supervised classification process. This paper introduced a classification network adopting a fuzzy-based $\gamma$-model in order to select a training data set and to extract feature which highly contribute to an actual classification. The features used in the classification were gray-level histogram, textures, and NDVI(Normalized Difference Vegetation Index) of target imagery. Moreover, in order to minimize the errors in the classification network, the Gradient Descent method was used in the training process for the $\gamma$-parameters at each code used. The trained parameters made it possible to know the connectivity of each node and to delete the void features from all the possible input features.

• Korean Journal of Cognitive Science
• /
• v.9 no.1
• /
• pp.1-11
• /
• 1998

In this paper, the features which are thought to be used by humans based on the psychological experiment of human information processing are applied to character recognition problem. Man will deal with a little large area information as well as pixel by pixel information. Therefore we define the feature that represents a little wide region I information called region feature, and combine the features derived from region feature and pixel by pixel features that have been used by now. The features we used are the result of region feature based preanalysis, mesh with region attributes, cross distance difference and gradient. The training and test data in the experiment are handwritten Korean alphabets, digits and English alphabets, which are trained on neural network using back propagation algorithm and recognition results are 90.27-93.25%, 98.00% and 79.73-85.75%, respectively Experimental results show that the feature we are suggesting in this paper is 1-2% better than UDLRH feature similar in attribute to region feature, and the tendency of misrecognition is more easily acceptable by humans.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.1-7
• /
• 2021

In this paper, we propose a feature selection technique for multi-label classification. Many existing feature selection techniques have selected features by calculating the relation between features and labels such as a mutual information scale. However, since the mutual information measure requires a joint probability, it is difficult to calculate the joint probability from an actual premise feature set. Therefore, it has the disadvantage that only a few features can be calculated and only local optimization is possible. Away from this regional optimization problem, we propose a feature selection technique that constructs a low-rank space in the entire given feature space and selects features with sparsity. To this end, we designed a regression-based objective function using Nuclear norm, and proposed an algorithm of gradient descent method to solve the optimization problem of this objective function. Based on the results of multi-label classification experiments on four data and three multi-label classification performance, the proposed methodology showed better performance than the existing feature selection technique. In addition, it was showed by experimental results that the performance change is insensitive even to the parameter value change of the proposed objective function.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1315-1318
• /
• 2003

In this paper, a neural network based approach using a self-organizing feature map is proposed for the segmentation of X ray images. A number of algorithms based on such approaches as histogram analysis, region growing, edge detection and pixel classification have been proposed for segmentation of general images. However, few approaches have been applied to X ray image segmentation because of blur of the X ray image and vagueness of its edge, which are inherent properties of X ray images. To this end, we develop a new model based on the neural network to detect objects in a given X ray image. The new model utilizes Mumford-Shah functional incorporating with a modified adaptive SOFM. Although Mumford-Shah model is an active contour model not based on the gradient of the image for finding edges in image, it has some limitation to accurately represent object images. To avoid this criticism, we utilize an adaptive self organizing feature map developed earlier by the authors.[1] It's learning rule is derived from Mumford-Shah energy function and the boundary of blurred and vague X ray image. The evolution of the neural network is shown to well segment and represent. To demonstrate the performance of the proposed method, segmentation of an industrial part is solved and the experimental results are discussed in detail.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.20 no.10
• /
• pp.3038-3047
• /
• 1996

In telerobot systems, visual servoing of a task object for a slave arm with an eye-in-hand has drawn an interesting attention. As such a task ingenerally conducted in an unstructured environment, it is very difficult to define the inverse feature Jacobian matrix. To overcome this difficulty, this paper proposes an auto-tuning fuzzy rule-based visual servo algorithm. In this algorithm, a visual servo controller composed of fuzzy rules, receives feature errors as inputs and generates the change of have position as outputs. The fuzzy rules are tuned by using steepest gradient method of the cost function, which is defined as a quadratic function of feature errors. Since the fuzzy rules are tuned automatically, this method can be applied to the visual servoing of a slave arm in real time. The effctiveness of the proposed algorithm is verified through a series of simulations and experiments. The results show that through the learning procedure, the slave arm and track object in real time with reasonable accuracy.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.3
• /
• pp.617-623
• /
• 2018

This study focuses on detection of malicious code through AndroidManifest permissoion feature extracted based on Android static analysis. Features are built on the permissions of AndroidManifest, which can save resources and time for analysis. Malicious app detection model consisted of SVM (support vector machine), NB (Naive Bayes), Gradient Boosting Classifier (GBC) and Logistic Regression model which learned 1,500 normal apps and 500 malicious apps and 98% detection rate. In addition, malicious app family identification is implemented by multi-classifiers model using algorithm SVM, GPC (Gaussian Process Classifier) and GBC (Gradient Boosting Classifier). The learned family identification machine learning model identified 92% of malicious app families.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.12 no.2
• /
• pp.199-205
• /
• 2009

A simple and effective analysis method is presented for predicting the best infiltration area. Based on geospatial information, numerical estimation barometer for degree of infiltration risk has been derived. The dominant geospatial features influencing infiltration risk have been found to be area altitude, degree of surface gradient, relative direction of surface gradient to the surveillance line, degree of surface gradient repetition, regional forest information. Each feature has been numerically expressed corresponding to the degree of infiltration risk of that area. Four different detection probability maps of infiltration risk for the surveillance area are drawn on the actual map with respect to the numerically expressed five dominant factors of infiltration risks. By combining the four detection probability maps, the complete picture of thr best infiltration area has been drawn. By using the map and the analytic method the effectiveness of surveillance operation can be improved.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.6
• /
• pp.492-503
• /
• 2013

H.264/AVC has shown high coding efficiency by using various coding tools, including intra and inter prediction. However, there are still many more redundancy components in intra prediction than in inter prediction. In this paper, a novel intra prediction method is proposed with adaptive mode selection. The combined intra prediction modes and simplified gradient modes are added in order to refine the directional feature and gradation region. Suitable modes are selected according to the neighboring blocks that provide a high compression rate and lower computational complexity. The improvement of the proposed method is 1.96% in terms of the bitrate, 0.25 dB in PSNR, and 1.72 times in terms of the computational complexity.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.10a
• /
• pp.382-383
• /
• 2016

Canny edge detection of the image is composed of four kinds of Gaussian filter, gradient calculation, Non-maximum suppression, and Hypothesis Thresholding. Feature is the ratio between the vehicle body, the windows, and the wheels obtained from the edge image. Features that make the proportion of these vehicles are different for each respective model. We have identified by application of this algorithm where only a small vehicle.