Bonhwa Ku

Performance comparison of target localization for active sonar systems

The performance comparison of target localization for active sonar and effective fusion algorithms for target localization of multistatic sonar is investigated. Active sonar can be categorized into monostatic, bistatic, and multistatic, depending on the number of receiver elements. Target localization performance depends on the system configuration. The target localization performance of the monostatic, bistatic, and multistatic sonar systems is compared assuming that each element can receive both range and azimuth information of the target. In addition, we propose the weighted least square (WLS) algorithm, which incorporates judicial weighting to the conventional least square (LS) method, and an efficient sensor arrangement rule for target localization in the multistatic sonar system. The representative experimental results demonstrate that the target localization performance of multistatic sonar configuration is superior in terms of root-mean-square error (RMSE), to monostatic sonar and bistatic sonar by 35.98% and 37.45%, respectively, while the proposed WLS algorithm showed an improvement of 2.27% compared with the LS method.

License Plate Detection Using Local Structure Patterns

We address the problem of license plate detection invideo surveillance systems. The Adaboost based approach,known for relative ease of implementation, makes use ofdiscriminative features such as edges or Haar-like features.In this paper, we propose a novel detection algorithm basedon local structure patterns for license plate detection. Theproposed algorithm includes post-processing methods toreduce false positive rate using positional and colorinformation of license plates. Experimental resultsdemonstrate effectiveness of the proposed methodcompared

Online multi-object tracking with efficient track drift and fragmentation handling

This paper addresses the problem of multi-object tracking in complex scenes by a single, static, uncalibrated camera. Tracking-by-detection is a widely used approach for multi-object tracking. Challenges still remain in complex scenes, however, when this approach has to deal with occlusions, unreliable detections (e.g., inaccurate position/size, false positives, or false negatives), and sudden object motion/appearance changes, among other issues. To handle these problems, this paper presents a novel online multi-object tracking method, which can be fully applied to real-time applications. First, an object tracking process based on frame-by-frame association with a novel affinity model and an appearance update that does not rely on online learning is proposed to effectively and rapidly assign detections to tracks. Second, a two-stage drift handling method with novel track confidence is proposed to correct drifting tracks caused by the abrupt motion change of objects under occlusion and prolonged inaccurate detections. In addition, a fragmentation handling method based on a track-to-track association is proposed to solve the problem in which an object trajectory is broken into several tracks due to long-term occlusions. Based on experimental results derived from challenging public data sets, the proposed method delivers an impressive performance compared with other state-of-the-art methods. Furthermore, additional performance analysis demonstrates the effect and usefulness of each component of the proposed method.

Online multi-person tracking with two-stage data association and online appearance model learning

This study addresses the automatic multi-person tracking problem in complex scenes from a single, static, uncalibrated camera. In contrast with offline tracking approaches, a novel online multi-person tracking method is proposed based on a sequential tracking-by-detection framework, which can be applied to real-time applications. A two-stage data association is first developed to handle the drifting targets stemming from occlusions and peoples abrupt motion changes. Subsequently, a novel online appearance learning is developed by using the incremental/decremental support vector machine with an adaptive training sample collection strategy to ensure reliable data association and rapid learning. Experimental results show the effectiveness and robustness of the proposed method while demonstrating its compatibility with real-time applications.

Simulation and Ship Detection Using Surface Radial Current Observing Compact HF Radar

This paper proposes an effective method of improving ship detection performance of a compact high-frequency (HF) radar system which has been primarily optimized for observing surface radial current velocities and bearings. Previously developed ship detection systems have been vulnerable to error sources such as environmental noise and clutter when they are applied in a compact HF radar optimized for observing surface current. In particular, the influences of error are reduced by applying a principle component analysis of the generated range-Doppler maps. A compact radar signal model is first developed by the data acquired from an operating compact HF radar site. The proposed method is then validated by comparing it to the conventional ship detection method in terms of detection and false alarm rates. The experimental results confirm that the proposed method shows superior performance in both simulated and practical environments.

Online multi-person tracking for intelligent video surveillance systems

This paper presents a novel online multi-person tracking method based on tracking-by-detection framework for intelligent video surveillance systems consisting of an internet protocol (IP) camera and a network video recorder (NVR). First, the two-step data association based on high/low confidence targets is proposed for handling long-term occlusions effectively. Additionally, the method that reasons and handles severe inter-target occlusion is proposed. Representative experimental results demonstrate the effectiveness and robustness of the proposed method.

Suppressing Ghost Targets via Gating and Track History in Y-Shaped Passive Linear Array Sonars

In Y-shaped passive linear array sonar (PLAS) systems composed of 3 sensor legs with consecutively missed detections in a full scan, the complex ghosts that occur due to bearing ambiguity seriously deteriorate the target tracking performance. A solution to the ghost problem is proposed which exploits the track information and geometrical relationship between each PLAS leg and the target. The region of possible ghost targets is first established by making use of the geometrical relationship of the PLAS with respect to the target. The ghost targets are then eliminated by means of the updated track information refined via a combination of 3-D assignment with gating and judicial track management. In the track management and gating procedure, we propose an optimal measurement selection criterion based on maximum joint probability data association (MJPDA), which combines both the soft and hard association techniques. The established tracks are updated by the Kalman filter with MJPDA and are terminated with a logic-based procedure. Representative simulations demonstrate the effectiveness of the proposed approach.

Robust video super resolution algorithm using measurement validation method and scene change detection

Explicit motion estimation is considered a major factor in the performance of classical motion-based super resolution (SR) algorithms. To reconstruct video frames sequentially, we applied a dynamic SR algorithm based on the Kalman recursive estimator. Our approach includes a novel measurement validation process to attain robust image reconstruction results under inexplicit motion estimation. In our method, the suitability for high-resolution pixel estimation is determined by the accuracy of motion estimation. We measured the accuracy of the image registration result using the Mahalanobis distance between the input low-resolution frame and the motion compensated high-resolution estimation. We also incorporate an effective scene change detection method dedicated to the proposed SR approach for minimizing erroneous results when abrupt scene changes occur in the video frames. According to the ratio of well-aligned pixels (i.e., motion is compensated accurately) to the total number of pixels, we are able to detect sudden changes of scene and context in the input video. Representative experiments on synthetic and real video data show robust performance of the proposed algorithm in terms of its reconstruction quality even with errors in the estimated motion.

Robust Dynamic Super Resolution under Inaccurate Motion Estimation

In image reconstruction, dynamic super resolutionimage reconstruction algorithms have been investigated toenhance video frames sequentially, where explicit motionestimation is considered as a major factor in theperformance. This paper proposes a novel measurementvalidation method to attain robust image reconstructionresults under inaccurate motion estimation. In addition, wepresent an effective scene change detection methoddedicated to the proposed super resolution technique forminimizing erroneous results when abrupt scene changesoccur in the video frames. Representative experimentalresults show excellent performance of the proposedalgorithm in terms of the reconstruction quality andprocessing speed.

Crowd Density Estimation Using Multi-class Adaboost

In this paper, we propose a crowd density estimation algorithm based on multi-class Adaboost using spectral texture features. Conventional methods based on self-organizing maps have shown unsatisfactory performance in practical scenarios, and in particular, they have exhibited abrupt degradation in performance under special conditions of crowd densities. In order to address these problems, we have developed a new training strategy by incorporating multi-class Adaboost with spectral texture features that represent a global texture pattern. According to the representative experimental results, the proposed method shows an average improvement of about 30% in the correct recognition rate, as compared to existing conventional methods.