Matches in SemOpenAlex for { <https://semopenalex.org/work/W3137165785> ?p ?o ?g. }
- W3137165785 endingPage "18" @default.
- W3137165785 startingPage "1" @default.
- W3137165785 abstract "Object detection is an essential task in computer vision. Recently, several convolution neural network (CNN)-based detectors have achieved a great success in natural scenes. However, for optical remote sensing images with a large scale of view, lower proportion of foreground target pixels and drastic differences in object scale present considerable challenges. To address these problems, we propose a novel one-stage detector called the full-scale object detection network (FSoD-Net) which consists of proposed multiscale enhancement network (MSE-Net) backbone cascaded with scale-invariant regression layers (SIRLs). First, MSE-Net provides the multiscale description enhancement by integrated the Laplace kernel with fewer parallel multiscale convolution layers. Second, SIRLs contain three different isolated regression branch layers (i.e., corresponding to small, medium, and large scales), which make default discrete scale bounding boxes (bboxes) cover full-scale object information in regression procedure. A novel specific scale joint loss is also designed that uses the softmax function combined with a strong <inline-formula> <tex-math notation=LaTeX>$L_{1}$ </tex-math></inline-formula>-norm constraint in each regression branch layer. It can further speed up the convergence and improve the classification scores of predicted bboxes. Finally, extensive experiments are carried on challenge data sets of large-scale dataset for object detection in aerial images (DOTA) and object detection in optical remote sensing images (DIOR) which contain multiple instances from different imaging platforms, and these results demonstrate that FSoD-Net can achieve better performance than other state-of-the-art one-stage detectors, and it can reach a mean average precision (mAP) of 75.33% on DOTA and 71.80% mAP on DIOR, respectively. Especially, the average precision (AP) of tiny object detection can improve 10%–20% approximately." @default.
- W3137165785 created "2021-03-29" @default.
- W3137165785 creator A5006719409 @default.
- W3137165785 creator A5023179941 @default.
- W3137165785 creator A5027713332 @default.
- W3137165785 creator A5043348210 @default.
- W3137165785 creator A5049266422 @default.
- W3137165785 creator A5056448190 @default.
- W3137165785 creator A5075134790 @default.
- W3137165785 creator A5029452373 @default.
- W3137165785 date "2022-01-01" @default.
- W3137165785 modified "2023-10-16" @default.
- W3137165785 title "FSoD-Net: Full-Scale Object Detection From Optical Remote Sensing Imagery" @default.
- W3137165785 cites W1904073808 @default.
- W3137165785 cites W2025242589 @default.
- W3137165785 cites W2029725672 @default.
- W3137165785 cites W2039967127 @default.
- W3137165785 cites W2085625911 @default.
- W3137165785 cites W2086355237 @default.
- W3137165785 cites W2097117768 @default.
- W3137165785 cites W2109255472 @default.
- W3137165785 cites W2112796928 @default.
- W3137165785 cites W2151103935 @default.
- W3137165785 cites W2161969291 @default.
- W3137165785 cites W2163808566 @default.
- W3137165785 cites W2194775991 @default.
- W3137165785 cites W2512351403 @default.
- W3137165785 cites W2549139847 @default.
- W3137165785 cites W2565639579 @default.
- W3137165785 cites W2577537809 @default.
- W3137165785 cites W2618530766 @default.
- W3137165785 cites W2733535455 @default.
- W3137165785 cites W2779335303 @default.
- W3137165785 cites W2787630273 @default.
- W3137165785 cites W2796045942 @default.
- W3137165785 cites W2800213945 @default.
- W3137165785 cites W2804436788 @default.
- W3137165785 cites W2884419823 @default.
- W3137165785 cites W2887362407 @default.
- W3137165785 cites W2888527098 @default.
- W3137165785 cites W2897898606 @default.
- W3137165785 cites W2899607431 @default.
- W3137165785 cites W2899836188 @default.
- W3137165785 cites W2913087080 @default.
- W3137165785 cites W2921357635 @default.
- W3137165785 cites W2928870406 @default.
- W3137165785 cites W2934276789 @default.
- W3137165785 cites W2962749812 @default.
- W3137165785 cites W2963037989 @default.
- W3137165785 cites W2963150697 @default.
- W3137165785 cites W2963351448 @default.
- W3137165785 cites W2963604034 @default.
- W3137165785 cites W2963857746 @default.
- W3137165785 cites W2969393175 @default.
- W3137165785 cites W2972256598 @default.
- W3137165785 cites W2989127086 @default.
- W3137165785 cites W2992240579 @default.
- W3137165785 cites W2994619704 @default.
- W3137165785 cites W2994928934 @default.
- W3137165785 cites W2998619030 @default.
- W3137165785 cites W3009818596 @default.
- W3137165785 cites W3009929949 @default.
- W3137165785 cites W3013696364 @default.
- W3137165785 cites W3034465353 @default.
- W3137165785 cites W3034971973 @default.
- W3137165785 cites W3098218837 @default.
- W3137165785 cites W3103856189 @default.
- W3137165785 cites W3106250896 @default.
- W3137165785 cites W3125372245 @default.
- W3137165785 doi "https://doi.org/10.1109/tgrs.2021.3064599" @default.
- W3137165785 hasPublicationYear "2022" @default.
- W3137165785 type Work @default.
- W3137165785 sameAs 3137165785 @default.
- W3137165785 citedByCount "17" @default.
- W3137165785 countsByYear W31371657852021 @default.
- W3137165785 countsByYear W31371657852022 @default.
- W3137165785 countsByYear W31371657852023 @default.
- W3137165785 crossrefType "journal-article" @default.
- W3137165785 hasAuthorship W3137165785A5006719409 @default.
- W3137165785 hasAuthorship W3137165785A5023179941 @default.
- W3137165785 hasAuthorship W3137165785A5027713332 @default.
- W3137165785 hasAuthorship W3137165785A5029452373 @default.
- W3137165785 hasAuthorship W3137165785A5043348210 @default.
- W3137165785 hasAuthorship W3137165785A5049266422 @default.
- W3137165785 hasAuthorship W3137165785A5056448190 @default.
- W3137165785 hasAuthorship W3137165785A5075134790 @default.
- W3137165785 hasConcept C121332964 @default.
- W3137165785 hasConcept C127313418 @default.
- W3137165785 hasConcept C153180895 @default.
- W3137165785 hasConcept C154945302 @default.
- W3137165785 hasConcept C160633673 @default.
- W3137165785 hasConcept C188441871 @default.
- W3137165785 hasConcept C2776151529 @default.
- W3137165785 hasConcept C2778755073 @default.
- W3137165785 hasConcept C31972630 @default.
- W3137165785 hasConcept C41008148 @default.
- W3137165785 hasConcept C62520636 @default.
- W3137165785 hasConcept C62649853 @default.