Synthetic aperture radar imaging of sea surface life and fishing activities

Petit, M.; Stretta, Jean-Michel; Farrugio, Henri; Wadsworth, A.

doi:10.1109/36.175346

Cited by 20 publications

(12 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(1) The detection accuracy of DS-CNN is slightly lower than C-CNN (94.13% mAP < 94.39% mAP). This phenomenon shows that using DS-CNN to build a network is feasible and does not decline the detection accuracy too much; (2) The detection speed of DS-CNN is 2.7 times than C-CNN (111 FPS > 41 FPS). This phenomenon shows that DS-CNN can indeed improve the detection speed by using D-Conv2D and P-Conv2D to replace C-CNN.…”

mentioning

confidence: 89%

“…From Equations (1) and (2), C-CNN has four convolution operations and DS-CNN has only one convolution operation, therefore, DS-CNN simplifies traditional convolution operation, which can obviously reduce the amount of calculation. To be clear, Equations (1) and (2) here are only rough expressions and more rigorous expressions will be introduced in Section 2.3. Figure 2 is the detailed internal operation flow of a DS-CNN.…”

Section: Ds-cnnmentioning

confidence: 99%

“…The detection accuracy is the highest when L = 160. The detection accuracy does not increase any more, but fluctuates slightly when L > 160; (2) The detection speed becomes lower with the increase of L (marked in pink in Figure 11b). This phenomenon is in line with common sense because large-size images have more pixels, leading to more computation.…”

mentioning

confidence: 95%

See 2 more Smart Citations

Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection

et al. 2019

View full text Add to dashboard Cite

As an active microwave imaging sensor for the high-resolution earth observation, synthetic aperture radar (SAR) has been extensively applied in military, agriculture, geology, ecology, oceanography, etc., due to its prominent advantages of all-weather and all-time working capacity. Especially, in the marine field, SAR can provide numerous high-quality services for fishery management, traffic control, sea-ice monitoring, marine environmental protection, etc. Among them, ship detection in SAR images has attracted more and more attention on account of the urgent requirements of maritime rescue and military strategy formulation. Nowadays, most researches are focusing on improving the ship detection accuracy, while the detection speed is frequently neglected, regardless of traditional feature extraction methods or modern deep learning (DL) methods. However, the high-speed SAR ship detection is of great practical value, because it can provide real-time maritime disaster rescue and emergency military planning. Therefore, in order to address this problem, we proposed a novel high-speed SAR ship detection approach by mainly using depthwise separable convolution neural network (DS-CNN). In this approach, we integrated multi-scale detection mechanism, concatenation mechanism and anchor box mechanism to establish a brand-new light-weight network architecture for the high-speed SAR ship detection. We used DS-CNN, which consists of a depthwise convolution (D-Conv2D) and a pointwise convolution (P-Conv2D), to substitute for the conventional convolution neural network (C-CNN). In this way, the number of network parameters gets obviously decreased, and the ship detection speed gets dramatically improved. We experimented on an open SAR ship detection dataset (SSDD) to validate the correctness and feasibility of the proposed method. To verify the strong migration capacity of our method, we also carried out actual ship detection on a wide-region large-size Sentinel-1 SAR image. Ultimately, under the same hardware platform with NVIDIA RTX2080Ti GPU, the experimental results indicated that the ship detection speed of our proposed method is faster than other methods, meanwhile the detection accuracy is only lightly sacrificed compared with the state-of-art object detectors. Our method has great application value in real-time maritime disaster rescue and emergency military planning.

show abstract

mentioning

confidence: 89%

Section: Ds-cnnmentioning

confidence: 99%

See 1 more Smart Citation

Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Ship detection with satellite based on Synthetic Aperture Radar (SAR) was first demonstrated by the experimental SEASAT in 1978. With later first–generation satellites such as ERS–1, JERS–1, ERS–2 the field has reached some maturity [2]. With the advent of the second generation of radar satellites such as ENVISAT and RADARSAT–1, ship detection capabilities were once and for all established thanks to advanced specific processing taking advantage of the huge amount of information that can be retrieved from low level products [3].…”

Section: Introductionmentioning

confidence: 99%

Using SPOT-5 HRG Data in Panchromatic Mode for Operational Detection of Small Ships in Tropical Area

Corbane

Marre

Petit

2008

Sensors

View full text Add to dashboard Cite

Nowadays, there is a growing interest in applications of space remote sensing systems for maritime surveillance which includes among others traffic surveillance, maritime security, illegal fisheries survey, oil discharge and sea pollution monitoring. Within the framework of several French and European projects, an algorithm for automatic ship detection from SPOT–5 HRG data was developed to complement existing fishery control measures, in particular the Vessel Monitoring System. The algorithm focused on feature–based analysis of satellite imagery. Genetic algorithms and Neural Networks were used to deal with the feature–borne information. Based on the described approach, a first prototype was designed to classify small targets such as shrimp boats and tested on panchromatic SPOT–5, 5–m resolution product taking into account the environmental and fishing context. The ability to detect shrimp boats with satisfactory detection rates is an indicator of the robustness of the algorithm. Still, the benchmark revealed problems related to increased false alarm rates on particular types of images with a high percentage of cloud cover and a sea cluttered background.

show abstract

“…Furthermore, SAR has been used to image sea surface life and fishing activities (Petit et al 1992). The application of these radars depends on the character of the pulse emitted, which may be long or short, of uniform frequency or of swept frequency.…”

Section: Radar Measurements Of Physical Ocean Featuresmentioning

confidence: 99%

Remote sensing of environmental indicators of potential fish aggregation: An overview

Klemas

2012

Baltica

View full text Add to dashboard Cite

Sustainable use of marine resources requires effective monitoring and management of the world’s fish stocks. Acoustic and electromagnetic remote sensing techniques are being used to help manage fisheries at sustainable levels, while also guiding fishing fleets to locate fish schools more efficiently. Fish tend to aggregate in ocean areas that exhibit conditions favored by specific fish species. Some of the relevant oceanographic conditions, such as sea surface temperature, ocean color (productivity) and oceanic fronts, which strongly influence natural fluctuations of fish stocks, can be observed and measured by remote sensors on satellites, aircraft and ships. The remotely sensed data are provided in near-real time to help fishermen save fuel and ship time during their search for fish, to modelers who produce fisheries forecasts, and to scientists who help develop strategies for sustainable fisheries management. This article describes how acoustic, optical and radar sensors on ships, satellites and aircraft are used with forecast models to improve the management and harvesting of fisheries resources.

show abstract

Synthetic aperture radar imaging of sea surface life and fishing activities

Cited by 20 publications

References 3 publications

Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection

Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection

Using SPOT-5 HRG Data in Panchromatic Mode for Operational Detection of Small Ships in Tropical Area

Remote sensing of environmental indicators of potential fish aggregation: An overview

Contact Info

Product

Resources

About