Detection Performance Of Detector Research Articles

Squeeze and Excitation Rank Faster R-CNN for Ship Detection in SAR Images

Synthetic aperture radar (SAR) ship detection is an important part of marine monitoring. With the development in computer vision, deep learning has been used for ship detection in SAR images such as the faster region-based convolutional neural network (R-CNN), single-shot multibox detector, and densely connected network. In SAR ship detection field, deep learning has much better detection performance than traditional methods on nearshore areas. This is because traditional methods need sea–land segmentation before detection, and inaccurate sea–land mask decreases its detection performance. Though current deep learning SAR ship detection methods still have many false detections in land areas, and some ships are missed in sea areas. In this letter, a new network architecture based on the faster R-CNN is proposed to further improve the detection performance by using squeeze and excitation mechanism. In order to improve performance, first, the feature maps are extracted and concatenated to obtain multiscale feature maps with ImageNet pretrained VGG network. After region of interest pooling, an encoding scale vector which has values between 0 and 1 is generated from subfeature maps. The scale vector is ranked, and only top $K$ values will be preserved. Other values will be set to 0. Then, the subfeature maps are recalibrated by this scale vector. The redundant subfeature maps will be suppressed by this operation, and the detection performance of detector can be improved. The experimental results based on Sentinel-1 images show that the detection performance of the proposed method achieves 0.836 which is 9.7% better than the state-of-the-art method when using F1 as matric and executes 14% faster.

Discussion on Test Method of Detectors for Flammable Gases

Abstract The paper discussed mainly two aspects when testing detection performances of detectors for flammable gases, measuring principle of sensors and application environment of detectors. For different kinds of sensors, the performance characteristics decide immunity from interference of temperature and other gas, and according to application field, some special tests, for example, EMC tests are necessary. And provided a clear thinking to combine design of tests with practical application of products, which made the detectors meet the needs of practical application. It also put forward some conceptions for future development of testing technology.

高分辨雷达扩展目标检测算法研究

The problem of extended target detection for high resolution radar is addressed. After modeling the radar echo signal, the optimal detector in the ideal case, in which all information of the target is known, is derived. The performance of the optimal detector, which gives an upper bound of the detection performance for all potential detectors, is analyzed. Then the GLRT detector is discussed in ve cases in which the information of the target is partially or entirely unknown. The detection performances of these detectors are analyzed, and the theoretical and simulation results con rm that the more information about the target a detector has, the better detection performance it shows. The order of the detection performance of these detectors reveals that the detection performance of the order accumulation detector gives an upper bound of the detection performance for all practical detector for which all the information of the target is completely unknown. A double threshold detector based on scatter number and GLRT(BS-GLRT) is proposed. In the proposed detector, the number of target scatters is estimated by comparing signal amplitudes at all range cells against the rst threshold, and all the amplitudes above the rst threshold are accumulated and compared against the second threshold to make detection decision. Simulation con rms that the proposed detector outperforms the existing methods in all cases and the detector performance is very close to the upper bound.

Role of new ohmic electrode metallization in detection performance of bulk semi-insulating GaAs radiation detectors

In this work, we concentrate on the study of the role of ohmic electrode metallization of radiation imaging detectors based on bulk semi-insulating (SI) GaAs. We fabricated SI GaAs radiation detectors with Ti/Pt/Au Schottky contact on one side and In/Au, Mg/Au and Gd/Au contacts on the other side of bulk material with the aim to form hole barrier Schottky contact with non-injecting effect. Current–voltage and noise characteristics and also detection performance of detectors are measured and evaluated. Observed results are discussed and compared with detector using standard AuGeNi non-alloyed ohmic metallization.