Yukyung Yang

Realistic infrared sequence generation by physics-based infrared target modeling for infrared search and track

Infrared search and track pursues the detection of sea-skimming infrared targets incoming from long distance. This paper presents a realistic synthetic target simulator for the development of infrared target detection and tracking algorithms. The proposed simulator consists of a 2-D background modeling part and a 3-D infrared target modeling part. Real infrared background images are used for the realistic modeling of background. Synthetic infrared target images are obtained by the consecutive processing of 3-D geometric modeling and radiometric modeling of targets according to target types, target distances, and atmospheric transmissivity. The experimental results validate the realistic modeling of the proposed method by comparing real observation sequence data.

Horizontal small target detection with cooperative background estimation and removal filters

Detecting small targets is essential for mitigating the sea-based Infrared search and track (IRST) problem. It is easy to detect small targets in homogeneous backgrounds such as the sky. When targets are on the border line of heterogeneous backgrounds such as the horizon in the sky and sea surface, solving the problem of detection becomes difficult. This paper presents a novel spatial filtering method, called Double Layered-Background Removal Filter (DL-BRF), for achieving high detection rates and low false alarm rates. DL-BRF consists of a Modified-Mean Subtraction Filter (M-MSF) and a consecutive Local-Directional Background Removal Filter (L-DBRF). M-MSF enhances the target signal and reduces background noise. L-DBRF removes horizontal structures, which upgrade the signal-to-clutter ratio and background suppression factor. L-DBRF used after M-MSF enhances the synergistic performance of horizontal target detection. We validate the superior performance of the proposed method via real evaluation tests.

Hybrid real-time tracking of non-rigid objects under occlusions

A mean shift algorithm has gained special attention in recent years due to its simplicity to enable real-time tracking. However, the traditional mean shift tracking algorithm can fail to track target under occlusions. In this paper we propose a novel technique which alleviates the limitation of mean shift tracking. Our algorithm employs the Kalman filter to estimate the target dynamics information. Moreover, the proposed algorithm performs the background check process to calculate the similarity which expresses how similar to target the background is. We then find the exact target position combining the motion estimation by Kalman filter and the color based estimation by the mean shift algorithm based on the similarity value. Therefore, the proposed algorithm can robustly track targets under several types of occlusion, while the mean shift and mean shift-Kalman filter algorithms fail.

Robust horizontal target detection with cooperative spatial filtering

A cooperative spatial filtering method is presented to detect small targets around horizontal region for infrared search and track (IRST). Double window filter (DWF) can enhance signal-to-noise ratio then directional background removal filter (DBRF) can subtract horizontal background structure. Experimental results present upgraded detection rate and false alarm rate.

Covariance-based band selection and its application to near-real-time hyperspectral target detection

Abstract. The matched filter (MF) and adaptive coherence estimator (ACE) show great effectiveness in hyperspectral target detection applications. Practical applications in which on-board processing is generally required demand real-time or near-real-time implementation of these detectors. However, a vast amount of hyperspectral data may make real-time or near-real-time implementation of the detection algorithms almost impossible. Band selection can be one of the solutions to this problem by reducing the number of spectral bands. We propose a new band selection method that prioritizes spectral bands based on their influence on the detection performance of the MF and ACE and discards the least influential bands. We validate the performance of our method using real hyperspectral images. We also demonstrate our technique on near-real-time detection tasks and show it to be a feasible approach to the tasks.

Image based analysis of atmosphere turbulence effect on the target detection

The performance of EO/IR systems is influenced on the testing circumstance. After studying theoretically the performance variations by environmental conditions, we estimate the practical performances of the systems. But the results of test in the field is beyond theoretical estimations due to sudden changes of the atmospheric conditions which becomes worse under low altitude and long distance between targets. So in this paper, we show distortions of target signatures by atmosphere turbulence using images from the field test for target detection, and also shows linear relations between the target detection rates and its theoretical expectations.

Image and sensor los-based automatic horizontal line detection and tracking for infrared search and track

It is important to stably extract horizon lines for sea-based infrared search and track. Most of the existing solutions for the problems only use single image to detect horizon line. Although this results in good accuracy for some images, it often fails to detect horizons in foggy, occluded environments. In this paper, we propose a novel horizon detection and tracking method that is robust to sensor vibrations and occlusions using image and sensor-based initial horizon detection and robust statistics. Local horizon optimization and tracking produce stable horizons in occluded environments. The experimental results validate the robustness of the proposed method in real infrared images.

The Measurement and Application of the Minimum Detectable Irradiance for the Infrared Point Source Detection System

A procedure and method for the MDI(Minimum Detectable Irradiance) measurement of an infrared point source detection system is described in detail and its experimental result is analyzed. The proposed measurement method for MDI can be realized with a collimator in the laboratory environment. In addition, an estimation method of the maximum detection range of the infrared point source detection system is introduced and its performed result is shown.

Robust Ordnance Flash Detection Based on Cooperative Temporal and Spatial Filters

In this paper, we propose a novel method which can detect ordnance firing events in IR images. The proposed algorithm is comprised of effective target detection and robust clutter rejection methods based on the temporalspatial cooperative filter. And additional clutter reduction is performed based on the proposed two features, NTFF (Number of Temporal Filter Frames) and SNS(Size-Normalized Signal). Experimental results show the effectiveness and feasibilities of our proposed algorithm.

Clutter reduction using track classification and clutter map generation

A clutter reduction method using track classification and clutter map generation for the real-time target detection and tracking system is presented. Track classification identifies background clutter tracks and this information is used to reduce false detection rate in the detection process. Experimental results show the proposed method can reduce considerable false track rate.