A New Method Based on Two-Stage Detection Mechanism for Detecting Ships in High-Resolution SAR Images

Xu, Yongli; Xiong, Wei; Lv, Yafei; Liu, Hengyan

doi:10.1051/matecconf/201712801014

Cited by 4 publications

(6 citation statements)

References 10 publications

(7 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The absolute values of the characteristic functions 0 ( ) and 1 ( ) are plotted in Figures 3(a) and (b), and the probability densities corresponding to them -0 ( ) and 1 ( )are drawn in Figures 4(a) and(b). Integration of the probability densities 0 ( ) and 1 ( ) in (13) shows that under the specified pulse power ( = 18 ), the false alarm and correct detection probabilities are = 10 −7 and = 0.96, respectively.…”

Section: Numerical Resultsmentioning

confidence: 99%

“…In fact, the exact calculation of ( 0 ) according to the Formula (15) under a high multiplicity of the poles requires cumbersome computations indeed. In this case, it is preferable to approximate calculations by the trace method [8] or by numerical integration of (9), (13), by means of MATLAB software [17].…”

Section: The Methods Of Detection Characteristics Calculatingmentioning

confidence: 99%

“…Dividing the total observation time T into independent synthesis intervals (that is used in common processing systems [9][10][11][12][13]15]) is substantiated with regard to a simpler processing implementation. However, in this case, multi-channel processing over the whole interval T is optimal such that it uses several weight functions that correspond to the eigenvalues of the processing matrix comparable to the maximum value.…”

Section: = +mentioning

confidence: 99%

“…calculating the probability density 0 ( ) according to the Formulas (11) and (12) or by numerical integration 0 ( ) according to the formula (9); calculating the probability ( 0 ) according to the Formulas (14) and (15) or by numerical integration of 0 ( ) according to the Formula (13), subsequently determining the threshold 0 , at which the required value of the false alarm probability F is achieved. b) In case the object is present calculating the definition matrix 1 = ( + + ), where is the correlation matrix of the object signal (depends on the area of the object, its orientation and position within the resolution element); calculating N eigenvalues of the matrix 1 ; calculating the characteristic function 1 ( ) according to the Formula (8); calculating the probability density 1 ( ) according to the Formulas (11) and (12) or by numerical integration of 1 ( ) according to the Formula (9); calculating the correct detection probability D according to the Formulas (14) and (15) or by numerical integration of 1 ( ) according to the Formula (13).…”

Section: Calculation Of the False Alarm And Correct Detection Probabimentioning

confidence: 99%

“…The problem is reduced to the formation of statistics providing the total power reflected by the element of the signal with the subsequent comparison of the found value with a certain selected threshold. Unlike previous studies [9][10][11][12][13], we consider the detection of an object as based not on the results of the secondary processing of its radar image, but on the comparison of the statistics generated at the SAR output with the selected threshold. Thus, the speed of making a decision on whether or not an object is present greatly increases.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

The Object Detection Efficiency in Synthetic Aperture Radar Systems

Chernoyarov

Dobrucký

Иванов

et al. 2020

IJE

View full text Add to dashboard Cite

The main purpose of this paper is to develop the method of characteristic functions for calculating the detection characteristics in the case of the object surrounded by rough surfaces. This method is to be implemented in synthetic aperture radar (SAR) systems using optimal resolution algorithms. By applying the specified technique, the expressions have been obtained for the false alarm and correct detection probabilities. In order to illustrate the effective application of the introduced approach in the case of the generic SAR system, the results are presented of the calculation of the detection characteristics of the signal from the extended object surrounded by a rough surface. It is shown that the analysis allows us to substantiate the structure of the SAR signal processing channel and to obtain the improved relations for the radio observation characteristics in this case. The efficiency of the optimal signal processing in SAR systems can also be determined without the approximate calculations involved.

show abstract

Section: Numerical Resultsmentioning

confidence: 99%

Section: The Methods Of Detection Characteristics Calculatingmentioning

confidence: 99%

Section: = +mentioning

confidence: 99%

Section: Calculation Of the False Alarm And Correct Detection Probabimentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The Object Detection Efficiency in Synthetic Aperture Radar Systems

Chernoyarov

Dobrucký

Иванов

et al. 2020

IJE

View full text Add to dashboard Cite

show abstract

A Deep-Learning-Based Lightweight Model for Ship Localizations in SAR Images

Bhattacharjee,

Shanmugam,

Das

2023

IEEE Access

View full text Add to dashboard Cite

Ship detection and localizing its position are indispensable in maritime surveillance and monitoring. Until early 2000, ship detection relied on human intelligence, but with the fast-processing speed, artificial intelligence (AI), especially deep learning, has replaced manual intervention with automatic localization in tracking naval activities. Taking advantage of the continuous and cloud-free ocean observations of Synthetic Aperture Radar (SAR), recent studies have demonstrated some success in utilizing SAR data to localize ships using deep-learning and other AI methods despite the accuracy of the models being lower than the acceptable limit. However, the existing models are inherently complex and time consuming in addition to demanding an extensive computational resource, which pose a significant challenge when applied to satellite-based data. This study presents a computationally efficient deep-learning-based algorithm that has a wider applicability and improves the accuracy over the existing models for ship localization in SAR images. Training and testing of this algorithm were conducted using the SAR Ship Dataset, which contains ship chips with complex backgrounds extracted from Gaofen-3 and Sentinel-1 satellite data. It produced the localized ship's position with bounding boxes in SAR images using the combined traditional computer vision and deep neural network configuration, which comprises a novel backbone network called Ship-Net or S-Net. The S-Net model has a thirteen-layer backbone feature extraction network and a four-layer regression network concatenated. Further, this study proposes a modified combined loss function for optimizing the model performance. A comparative analysis of the proposed S-Net model was done using the various pre-build model architectures and loss function combinations. The results showed that the S-Net model with a combined loss function yielded 94.88% precision and 79.68% recall, with 12.58% precision and 7.39% recall higher than the state-of-the-art Faster RCNN baseline model. The proposed S-Net model has a relatively higher performance than the existing state-of-the-art models for ship localization in SAR images and can become an efficient tool for ship localization in optical images with suitable architectural and training scheme modifications for better coastal surveillance and worldwide naval security.

show abstract