A post-processing scheme for the performance improvement of vehicle detection in wide-area aerial imagery

Original Paper


In this paper, we present a post-processing scheme to improve the performance of vehicle detection in wide-area aerial imagery. Using low-resolution aerial frames for the performance analysis, we adapted nine algorithms for vehicle detection. We derived a three-stage scheme to measure performance improvement on the selected five object segmentation algorithms before and after post-processing. We compared automatic detections results to ground-truth objects, and classified each type of detections in terms of true positive, false negative and false positive. Several evaluation metrics are adopted for the experimental study.


Vehicle detection Segmentation Deep learning Post-processing Aerial imagery Overlap 



The author declares no conflict of interests on research. The author owes special gratitude to anonymous reviewers for their valuable comments on improving technical quality of this manuscript.


  1. 1.
    Aspiras, T.H., Asari, V.K., Vasquez, J.: Gaussian ringlet intensity distribution (GRID) features for rotation-invariant object detection in wide area motion imagery. In: 21st IEEE International Conference on Image Process (ICIP), pp. 2309–2313 (2014)Google Scholar
  2. 2.
    Philip, R.C., Ram, S., Gao, X., Rodríguez, J.J.: A comparison of tracking algorithm performance for objects in wide area imagery. In: IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI), pp. 109–112 (2014)Google Scholar
  3. 3.
    Teutsch, M.: Moving object detection and segmentation for remote aerial video surveillance. Ph.D. dissertation, Karlsruhe Institute of Technology, Germany (2014)Google Scholar
  4. 4.
    Türmer, S.: Car detection in low-frame rate aerial imagery of dense urban areas. Ph.D. dissertation, Technische Univ. ät München (2014)Google Scholar
  5. 5.
    Gao, X., Ram, S., Rodríguez, J.J.: A performance comparison of automatic detection schemes in wide-area aerial imagery. In: IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI), pp. 125–128 (2016)Google Scholar
  6. 6.
    Gao, X.: A thresholding scheme of eliminating false detections on vehicles in wide-area aerial imagery. Int. J. Signal Imaging Syst. Eng. 11(4), 217–224 (2018)CrossRefGoogle Scholar
  7. 7.
    Gleason, J., Nefian, A.V., Bouyssounousse, X., Fong, T., Bebis, G.: Vehicle detection from aerial imagery. In: IEEE International Conference on Robotics and Automation (ICRA), pp. 2065–2070 (2011)Google Scholar
  8. 8.
    Hou, X.-D., Zhang, L.-Q.: Saliency detection: a spectral residual approach. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1–8 (2007)Google Scholar
  9. 9.
    Huang, Z.-H., Leng, J.-S.: Texture extraction in natural scenes using region-based method. J. Digit Inf. Manag. 12(4), 246–254 (2014)Google Scholar
  10. 10.
    Jain, I., Rani, B.: Vehicle detection using image processing and fuzzy logic. Int. J. Comput. Sci. Commun. 1(2), 255–257 (2010)Google Scholar
  11. 11.
    Davies, E.R.: The Laplacian operator. In: Davies, E.R. (ed.) Computer and Machine Vision: Theory, Algorithms, Practicalities, pp. 131–134. Academic Press, Waltham (2012)Google Scholar
  12. 12.
    Saha, N.B., Ray, N.: Image thresholding by variational minimax optimization. Pattern Recognit 42(5), 843–856 (2009)CrossRefGoogle Scholar
  13. 13.
    Seo, N., Matlab computer vision and pattern recognition toolbox. (2007)
  14. 14.
    Mallat, S.: Wavelet transform modulus maxima; multiscale edge detection. In: Mallat, R. (ed.) A Wavelet Tour of Signal Processing: The Sparse Way, 3rd edn., pp. 218–240. Academic Press (2008)Google Scholar
  15. 15.
    Trujillo-Pino, A., Krissian, K., Alemán-Flores, M., Santana-Cedrés, D.: Accurate subpixel edge location based on partial area effect. Image Vis. Comput. 31(1), 72–90 (2013)CrossRefGoogle Scholar
  16. 16.
    Zheng, Z.-Z., Zhou, G.-Q., Wang, Y., Liu, Y.-L., Li, X.-W., Wang, X.-T., Jiang, L.: A novel vehicle detection method with high resolution highway aerial image. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 6(6), 2343–2348 (2013)CrossRefGoogle Scholar
  17. 17.
    Shaikh, S.H., Saeed, K., Chaki, N.: Evaluation measures. In: Shaikh, S.H., Saeed, K., Chaki, N. (eds.) Moving Object Detection using Background Subtraction, pp. 30–31. Springer (2014)Google Scholar
  18. 18.
    Sun, J.-G., Lu, H.-C., Liu, X.-P.: Saliency region detection based on Markov absorption probabilities. IEEE Trans. Image Process. 24(5), 1639–1649 (2015)MathSciNetCrossRefGoogle Scholar
  19. 19.
    Li, S., Zhou, G.-Q., Zheng, Z.-Z., Liu, Y.-L., Li, X.-W., Zhang, Y., Yue, T.: The relation between accuracy and size of structure element for vehicle detection with high resolution highway aerial images. In: IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pp. 2645–2648 (2013)Google Scholar
  20. 20.
    Borji, A., Cheng, M.-M., Jiang, H.-Z., Li, J.: Salient object detection: a benchmark. IEEE Trans. Image Process. 24(12), 5706–5722 (2015)MathSciNetCrossRefGoogle Scholar
  21. 21.
    Otsu, N.: A threshold selection method from gray-level histogram. IEEE Trans. Syst. Man Cybern. 9(1), 62–66 (1979)CrossRefGoogle Scholar
  22. 22.
    Marques, O.: Morphological Image Processing. Practical Image and Video Processing Using MATLAB, pp. 299–334. Wiley, New York (2011)CrossRefGoogle Scholar
  23. 23.
    Salem, M.A., Ghamry, N., Meffert, B.: Daubechies versus biorthogonal wavelets for moving object detection in traffic monitoring systems. Inform. Ber. 229, 8–9 (2009)Google Scholar
  24. 24.
    Samarabandu, J., Liu, X.-Q.: An edge-based text region extraction algorithm for indoor mobile robot navigation. Int. J. Signal Process. 3(4), 273–280 (2007)Google Scholar
  25. 25.
    Sharma, B., Katiyar, V.K., Gupta, A.K., Singh, A.: The automated vehicle detection of highway traffic images by differential morphological profile. J. Transp. Technol. 4, 150–156 (2014)CrossRefGoogle Scholar
  26. 26.
    Nascimento, J.C., Marques, J.S.: Performance evaluation of object detection algorithms for video surveillance. IEEE Trans. Multimed. 8(4), 761–774 (2006)CrossRefGoogle Scholar
  27. 27.
    Boughorbel, S., Jarray, F., El-Anbari, M.: Optimal classifier for imbalanced data using Matthews correlation coefficient metric. PLoS One 12(6), e0177678 (2017)CrossRefGoogle Scholar
  28. 28.
    Liu, Y., Cheng, J., Yan, C., Wu, X., Chen, F.: Research on the Matthews correlation coefficients metrics of personalized recommendation algorithm evaluation. Int. J. Hybrid Inf. Technol. 8(1), 163–172 (2015)CrossRefGoogle Scholar
  29. 29.
    Kurz, F., Azimi, S.-M., Sheu, C.-Y., D’Angelo, P.: Deep learning segmentation and 3D reconstruction of road markings using multiview aerial imagery. ISPRS Int. J. Geo Inf. 8(47), 1–16 (2019)Google Scholar
  30. 30.
    Tang, T.-Y., Zhou, S.-L., Deng, Z.-P., Lei, L., Zou, H.-X.: Arbitrary-oriented vehicle detection in aerial imagery with single convolutional neural networks. Remote Sens. 9(11), 1170 (2017). (1–17) CrossRefGoogle Scholar
  31. 31.
    Vasu, B.-K.: Visualizing resiliency of deep convolutional network interpretations for aerial imagery. Master’s thesis, Rochester Institute of Technology (2018)Google Scholar
  32. 32.
    Wang, Z.-B., Liu, K., Li, J., Zhu, Y., Zhang, Y.-N.: Various frameworks and libraries of machine learning and deep learning: a survey. Arch. Comput. Methods Eng. (2019). CrossRefGoogle Scholar
  33. 33.
    Yang, Y.M., Liao, W.-T., Li, X.-B., Rosenhahn, B.: Deep learning for vehicle detection in aerial images. In: 25th IEEE International Conference on Image Processing (ICIP), pp. 3079–3083 (2018)Google Scholar
  34. 34.
    Porter, R., Fraser, A.M., Hush, D.: Wide-area motion imagery. IEEE Signal Process. Mag. 27(5), 56–65 (2010)CrossRefGoogle Scholar
  35. 35.
    Zhong, J.-D., Lei, T., Yao, G.-L.: Robust vehicle detection in aerial images based on cascaded convolutional neural networks. Sensors 17(12), 2721 (2017). (1–18) CrossRefGoogle Scholar
  36. 36.
    Zhu, H.-G.: Research of high-resolution aerial object detection based on deep learning. Master’s thesis, University of Chinese Academy of Science, China (2015)Google Scholar
  37. 37.
    Zhu, X.-Z., Dai, J.-F., Yuan, L., Wei, Y.-C.: Towards high performance video object detection. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 7210–7218 (2018)Google Scholar
  38. 38.
    Liu, X., Yang, T., Li, J.: Real-time ground vehicle detection in aerial infrared imagery based on convolutional neural network. Electronics 7(6), 78 (2018)CrossRefGoogle Scholar
  39. 39.
    Shao, S.-C, Tunc, C., Satam, P., Hariri, S.: Real-time IRC threat detection framework. In: IEEE 2nd International Workshops on Foundation and Applications of Self* Systems (FAS* W), pp. 318–323 (2017)Google Scholar
  40. 40.
    Gao, X.: Automatic Vehicle Detection, Segmentation and Tracking in Wide-Area Aerial Imagery. Thesis, The University of Arizona, Tucson, USA (2016)Google Scholar
  41. 41.
    Gao, X.: Vehicle detection in wide-area aerial imagery: cross-association of detection schemes with post-processings. Int. J. Image Min. 3(2), 106–116 (2018)CrossRefGoogle Scholar
  42. 42.
    Yang, B., Zhang, S., Tian, Y., Li, B.-J.: Front-vehicle detection in video images based on temporal and spatial characteristics. Sensors 19(7), 1728 (2019). CrossRefGoogle Scholar
  43. 43.
    Zhang, X.-F., Feng, G.-P., Gao, X., Xu, D.-Z.: Blind multiuser detection for MC-CDMA with antenna array. Comput. Elect. Eng. 36(1), 160–168 (2010)CrossRefGoogle Scholar
  44. 44.
    Pacheco, J., Satam, S., Hariri, S., Grijalva, C., Berkenbrock, H.: IoT security development framework for building trustworthy smart car devices. In: IEEE Conference on Intelligence and Security Informatics (ISI), pp. 237–242 (2016)Google Scholar
  45. 45.
    Bernard, J., Shao, S.-C, Tunc, C., Kheddouci, H., Hariri, S.: Quasi-cliques analysis for IRC channel thread detection. In: International Conference on Complex Networks and their Applications, pp. 578–589. Springer, Cham (2018)Google Scholar

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Electrical and Computer EngineeringThe University of ArizonaTucsonUSA

Personalised recommendations