Metrics for Performance Evaluation of Preprocessing Algorithms in Infrared Small Target Images

Diao, Wenhui; Mao, Xia; Gui, Vasile

doi:10.2528/pier11012412

Cited by 9 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If there is a one-pixel target in the IR image, IT represents the target [10]. If the target consists of more than one pixel, IT represents the minimum [18], maximum [19], or mean [20] intensity value of the target. In the literature, the mean value is mostly used because it represents all pixel values of the target [20].…”

Section: Existing Snr Metricsmentioning

confidence: 99%

“…If the target consists of more than one pixel, IT represents the minimum [18], maximum [19], or mean [20] intensity value of the target. In the literature, the mean value is mostly used because it represents all pixel values of the target [20]. However, in case of bimodal target, i.e., if some parts of the target are brighter and other parts are darker than the background, intensity of the target IT becomes low because of the mean calculation.…”

Section: Existing Snr Metricsmentioning

confidence: 99%

See 1 more Smart Citation

Pixel-Wise Signal-to-Noise Ratio: A Novel Metric for Quantifying the Detectability of Targets in Infrared Images

Tunç¹,

Ilgin²

2023

View full text Add to dashboard Cite

In this paper, a new Signal-to-Noise Ratio (SNR) metric is proposed to quantify the detectability of targets in infrared (IR) images. The proposed metric is based on the contrast between the target and the background, which is consistent with human perception in terms of distinguishing the target from the background, rather than the raw intensity values of the target. In the contrast calculation, individual contribution of each pixel value of the target is considered in the proposed metric, whereas the mean or a single representative raw intensity value of the target is taken into account in the existing metrics. As subjective evaluations are the most precise tools for distinguishing the target from the background, SNR metrics used for IR images are expected to be as consistent as possible with the human visual system. That is, due to its high contrast sensitivity, the human visual system responds to stimuli by cognitively distinguishing the target from the background. Therefore, human perceptioninspired target distinguishability metrics aim to quantify the target detectability consistent with the human visual system, which is capable of distinguishing very small differences in contrast. Extensive performance evaluation tests on well-known IR image datasets, VIVID, SENSIAC and AMCOM, and synthetic image sets demonstrate that the proposed pixel-wise SNR metric quantifies target distinguishability from the background more consistently with subjective evaluations than other SNR metrics. Furthermore, the proposed metric is always robust even when the other metrics fail to accurately quantify target distinguishability.

show abstract

Section: Existing Snr Metricsmentioning

confidence: 99%

Section: Existing Snr Metricsmentioning

confidence: 99%

Pixel-Wise Signal-to-Noise Ratio: A Novel Metric for Quantifying the Detectability of Targets in Infrared Images

Tunç¹,

Ilgin²

2023

View full text Add to dashboard Cite

show abstract

“…Image sequence-based motion detection technologies are widely applied in a variety of practical surveillance applications [1]. Traditional target detection on single frame image generally makes use of imaging characteristics of the targets of interest, such as in synthetic aperture radar (SAR) images [2,3] and infrared images [4,5]. However, for moving target detection from image sequences, the target motion is the most distinguishable characteristic to effectively extract targets.…”

Section: Introductionmentioning

confidence: 99%

Visual Image Sequential Motion Detection via Half Quadratic Minimization Method

Zhu¹,

Yun-li²,

An³

2018

PIER M

View full text Add to dashboard Cite

In this paper, we present a straightforward numerical algorithm for visual image sequential motion detection based on half quadratic minimization method. As for the standard visual image sequences with RGB color representation, an intuitive way is to convert it to grayscale image to achieve an approximate motion detection with relatively low computational load. Instead, we propose a sequential processing scheme for more accurate detection by utilizing the motion detection algorithm separately and then performing fusion on a higher level. In this way, we extend the motion detection technique for grayscale images to be capable of dealing with visual videos. Experiment results show that the proposed algorithm can provide more robust motion detection performance and be successfully utilized in practical visual surveillance applications.

show abstract

“…There are many radar signatures which can be used for radar target recognition: Natural frequencies of radar target [159][160][161][162][163][164][165][166], high resolution range (HRR) profiles [72,[75][76][77][78][79][80] of radar target, microwave image of the radar target [1,26,31,50,55,70,73,[81][82][83][84][85][86][87][88][89][90][91][92][93][94][95][96][97][98][99][100], and inverse synthetic aperture radar (ISAR) [55, 86-100, 125, 126] image proved to be useful features for target recognition. Jet engine modulation and helicopter modulation [127,128] have also been known as useful features for target recognition.…”

Section: Introductionmentioning

confidence: 99%

Effect of Threshold Value on the Performance of Natural Frequency-Based Radar Target Recognition

Cho¹,

Lee²

2013

PIER

View full text Add to dashboard Cite

Abstract-In this paper, the performance analysis of the natural frequency-based radar target recognition in the time domain is considered. We investigate the dependence of the probability of correct classification on a specific threshold value, and determine the optimum threshold value for two targets, and the sub-optimal threshold for multiple targets to maximize the probability of correct classification. Based on the probability density function (PDF) of some quantity consisting of the projections of the late time response onto the column spaces of the matrices constructed using the natural frequencies of the specific targets, we propose how to determine an optimum threshold in the sense that the probability of correct classification of two targets is maximized. By extending the scheme for two targets, we show how to determine a threshold value close to the optimal threshold for multiple targets. The scheme is validated by comparing the performance using the analytic method with that using the Monte-Carlo simulation.

show abstract

Metrics for Performance Evaluation of Preprocessing Algorithms in Infrared Small Target Images

Cited by 9 publications

References 26 publications

Pixel-Wise Signal-to-Noise Ratio: A Novel Metric for Quantifying the Detectability of Targets in Infrared Images

Pixel-Wise Signal-to-Noise Ratio: A Novel Metric for Quantifying the Detectability of Targets in Infrared Images

Visual Image Sequential Motion Detection via Half Quadratic Minimization Method

Effect of Threshold Value on the Performance of Natural Frequency-Based Radar Target Recognition

Contact Info

Product

Resources

About