Paper Title : OBJECT DETECTION USING SINGLE SHOT MULTIBOX DETECTOR (SSD)

ISSN : 2394-2231
Year of Publication : 2022

10.5281/zenodo.6387617

Authors: KAMAL CHANDWANI, VEDITA JANBANDHU, ANJALI SELOKAR, YASHKIRAN YERPUDE, KUNAL KAKDE, HASAN MESHRAM

         



MLA Style: OBJECT DETECTION USING SINGLE SHOT MULTIBOX DETECTOR (SSD) " KAMAL CHANDWANI, VEDITA JANBANDHU, ANJALI SELOKAR, YASHKIRAN YERPUDE, KUNAL KAKDE, HASAN MESHRAM " Volume 9 - Issue 2 International Journal of Computer Techniques (IJCT) ,ISSN:2394-2231 , www.ijctjournal.org

APA Style: OBJECT DETECTION USING SINGLE SHOT MULTIBOX DETECTOR (SSD) " KAMAL CHANDWANI, VEDITA JANBANDHU, ANJALI SELOKAR, YASHKIRAN YERPUDE, KUNAL KAKDE, HASAN MESHRAM " Volume 9 - Issue 2 International Journal of Computer Techniques (IJCT) ,ISSN:2394-2231 , www.ijctjournal.org

Abstract
In order to improve the detection accuracy of objects at different scales, the most recent studies applied multilayer architecture. However, the extracted low-level feature in the shallow layers may not work perfectly on the detection performance due to its less semantic information, especially for small objects. In this paper, we propose a refined featurefusion structure to be integrated with single shot detector (SSD). To obtain the rich representation ability for feature mapping, in the fusion block, the deconvolution operation is basically applied to fuse high-level semantic features and lowlevel semantic features. It is noteworthy that in the proposed framework, the feature pyramid network is modified to better describe the features by the skip connection. An adaptive weighted connection is designed at the feature-fusion block, which further enhances the performance of the detection. On PASCAL VOC2007 test set, the experimental results show that the map of the proposed network is higher than SSD and deconvolutional single shot detector (DSSD) by 2.03% and 0.63%, respectively. Meanwhile, the speed of our method is as 2.2 times fast as the DSSD. Furthermore, the map of our refined feature-fusion structure SSD is 6.2% higher than SSD on the small object test set of PASCALS VOC2007, which verifies the effectiveness of the proposed model.

Reference
[1] Y. Zhong, Y. Yang, X. Zhu, E. Dutkiewicz, Z. Zhou, T. Jiang, Device-free sensing for personnel detection in a foliage environment. IEEE Geoscience and Remote Sensing Letters 14(6), 921–925 (2017). https://doi.org/10.1109/LGRS.2017.268793 8 [2] S.Z. Su, S.Z. Li, S.Y. Chen, G.R. Cai, Y.D. Wu, A survey on pedestrian detection. DianziXuebao 40(4), 814–820 (2012). https://doi.org/10.3969/j.issn.0372- 2112.2012.04.031 [3] M. Zeng, J. Li, Z. Peng, The design of top-hat morphological filter and application to infrared target detection. Infrared Physics & Technology 48(1), 67–76 (2006). https://doi.org/10.1016/j.infrared.2005.04.0 06 [4] W. Liu, D. Angelo, D. Erhan, C. Szeged, S. Reed, C.Y. Fu, A.C. Berg, SSD: Single shot multibox detector. In European conference on computer vision (pp. 21- 37)(2016, October). Springer. Cham.. https://doi.org/10.1007/978-3-319-46448-0_2 [5] Z. Li, F. Zhou, FSSD: feature fusion single shot multibox detector. arXiv preprint arXiv:1712.00960.(2017). [6] J. Jeong, H. Park, N. Kwak, Enhancement of SSD by concatenating feature maps for object detection. arXiv preprint arXiv:1705.09587.(2017) [7] J.Q. Wang,J.S. Li,X.W. Zhou,X. Zhang, Improved SSD algorithm and its performance analysis of small target detection in remote sensing images[J]. Acta Optica Sinica,0628005(2019).https://doi.org/10.3788/AOS201939.062 8005

Keywords
— Python, CNN, Image processing, Datasets.