Inception-SSD: An improved single shot detector for vehicle detection


Vehicle detection plays an effective and important role in traffic safety, which has attracted extensive attention from both academic and industry. Deep learning has made significant breakthroughs in vehicle detection application. The Single Shot Detector (SSD) algorithm, which is one of the object detection algorithms, is used to detect vehicles. However, its main challenge is that the computing complexity and low accuracy. In this paper, an improved vehicle detection algorithm based on SSD is proposed to improve accuracy, especially for small vehicles detection. We add an Inception block to the extra layer in the SSD before the prediction to improve its performance. Then we use a new method that is more suitable for vehicle detection to set the scales and aspect ratios of the default bounding boxes, which benefits position regression and maintains the fast speed. The validity of our algorithm is verified on KITTI and UVD datasets. Compared with SSD, our algorithm achieves a higher mean average precision (mAP), while maintaining a fast speed.

This is a preview of subscription content, log in to check access.

Fig. 2
Fig. 3
Fig. 5
Fig. 6
Fig. 7
Fig. 8


  1. Piao J, Mcdonald M (2008) Advanced driver assistance systems from autonomous to cooperative approach. Transp Rev 28(5):684–695

    Article  Google Scholar 

  2. Mita T, Kaneko T, Hori O (2005) Joint haar-like features for face detection. Tenth IEEE Int Conf Computer Vision 2:1619–1626

    Article  Google Scholar 

  3. Ma X, Grimson WEL (2005) Edge-based rich representation for vehicle classification. Tenth IEEE Int Conf Computer Vision 2:1185–1192

    Google Scholar 

  4. Candes EJ, Tao T (2006) Near-optimal signal recovery from random projections: Universal encoding strategies? IEEE Trans Inf Theory 52(12):5406–5425

    MathSciNet  Article  Google Scholar 

  5. Lan R, Lu H, Zhou Y et al (2019) An lbp encoding scheme jointly using quaternionic representation and angular information. Neural Comp Appl 1–7

  6. Kazemi FM , Samadi S , Pourreza HR et al (2007) Vehicle recognition using curvelet transform and svm. Int Conf Inform Technol 516–521

  7. Freund Y, Schapire RE (1997) A desicion-theoretic generalization of on-line learning and an application to boosting. Comput Syst 55(1):119–139

    Article  Google Scholar 

  8. Lan R, Zhou Y, Liu Z et al (2018) Prior knowledge-based probabilistic collaborative representation for visual recognition. IEEE Trans Cybernet 50(4):1498–1508

    Article  Google Scholar 

  9. Chen, Shuhan et al (2018) Reverse Attention for Salient Object Detection. Eur Conf Computer Vision 236–52

  10. Liu W, Anguelov D, Erhan D et al (2016) SSD: Single shot multibox detector. European conference on computer vision 21–37

  11. Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. Int Conf Machine Learn 448–456

  12. Ekrem B, Altun Y (2017) Classification of vehicles in traffic and detection faulty vehicles by using ann techniques. Electric Electron Computer Sci.

    Article  Google Scholar 

  13. Druzhkov PN, Kustikova VD (2016) A survey of deep learning methods and software tools for image classification and object detection. Pattern Recogn Image Analy 26(1):9–15

    Article  Google Scholar 

  14. Zhou X, Wei G, Fu WL, Du F (2017) Application of deep learning in object detection. Int Conf Computer Inform Sci 631–634

  15. Serikawa S, Hui L (2014) Underwater image dehazing using joint trilateral filter. Comput Electr Eng 40(1):41–50

    Article  Google Scholar 

  16. Girshick R, Donahue J, Darrell T et al (2014) Rich feature hierarchies for accurate object detection and semantic segmentation. Computer Vision Pattern Recogn 580–587

  17. Girshick R. (2015) Fast r-cnn. Int Conf Computer Vision 1440–1448

  18. Ren S, He K, Girshick R et al (2015) Faster r-cnn : towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Machine Intell 39(6):1137–1149

    Article  Google Scholar 

  19. Redmon J, Divvala S, Girshick R et al (2016) You only look once: unified, real-time object detection. Computer Vision Pattern Recogn 779–788

  20. Karen S, Andrew Z (2015) Very deep convolutional networks for large-scale image recognition. Int Conf Learn Represent 1–14

  21. Redmon J, Farhadi A (2017) Yolo9000: better, faster, stronger. Computer Vis Pattern Recogn 6517–6525

  22. Thorson IL, Liénard J, David SV (2015) The essential complexity of auditory receptive fields. PLoS Comput Biol.

    Article  Google Scholar 

  23. Ning N C, Zhou N H, Song N Y et al (2017) Inception single shot multibox detector for object detection. IEEE Int Conf Multimedia Expo Workshops 549–554

  24. Hartigan JA, Wong MA (1979) Algorithm as 136: A k-means clustering algorithm. J Royal Statist Soc 28(1):100–108

    MATH  Google Scholar 

  25. Geiger A, Lenz P, Stiller C et al (2013) Vision meets robotics: The kitti dataset. Int J Robot Res 32(11):1231–1237

    Article  Google Scholar 

  26. Ranjeeth KC, Anuradha R (2020) Feature selection and classification methods for vehicle tracking and detection. J Amb Intell Human Comput.

    Article  Google Scholar 

  27. Armengol E (2019) Constructing a classifier with patterns. J Amb Intell HumanComput.

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Wanpei Chen.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Chen, W., Qiao, Y. & Li, Y. Inception-SSD: An improved single shot detector for vehicle detection. J Ambient Intell Human Comput (2020).

Download citation


  • Vehicle detection
  • Deep learning
  • Computer vision
  • Deep neural network
  • SSD