Advertisement

Fusion CNN Based on Feature Selection for Crime Scene Investigation Image Classification

  • Qiannan ZhangEmail author
  • Ying Liu
  • Fuping Wang
  • Jin Lu
  • Daxiang Li
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1075)

Abstract

Crime Scene Investigation images have many semantic categories and complex image contents. The Convolution Neural Network (CNN) feature cannot express the uniformity of image content and high dimensional features can lead to redundancy of feature vectors in CNN. In the circumstance it is difficult to use CNN to process crime scene investigation images. To solve the above problems, we propose a fusion CNN algorithm based on feature selection for the classification of crime scene investigation images. In this paper, we build the fusion CNN features to enhance the ability of representation by fusing the convolutional layer with the fully connected layer. Then we select the fusion features with Laplacian score and label mutual information. Finally, we use the obtained features to train Support Vector Machine (SVM) classifier on the Crime Scene Investigation Images Database (CSID). Experiments show that the average classification accuracy of the proposed method can reach 93.67%.

Keywords

Crime Scene Investigation Images classification Convolutional Neural Network Transfer learning Feature selection 

Notes

Acknowledgments

This work was supported by Project of International Science and Technology Cooperation and Exchange in Shaanxi Province of China (2018KW-003), National Natural Science Fund of China (61802305), and the graduate innovation fund project of Xi’an University of Posts and Telecommunications (CXJJLI2018012).

References

  1. 1.
    Liu, Y., Hu, D., Fan, J.L.: A survey of crime scene investigation image retrieval. Acta Electronica Sinica 46(3), 761–768 (2018)Google Scholar
  2. 2.
    Zhao, Y.D., Wang, Q., Fan, J.L.: Fuzzy KNN classification for criminal investigation image scene. Appl. Res. Comput. 31(10), 3158–3160 (2014)Google Scholar
  3. 3.
    Liu, Y., Yan, H., Lim, K.P.: Study on rotation-invariant texture feature extraction for tire pattern retrieval. Multidimension. Syst. Signal Process. 28(2), 757–770 (2017)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Bai, X.J., Shi, T.Y., Liu, Y.: An improved criminal scene investigation image classification algorithm. J. Xi’an Univ. Posts Telecommun. 21(6), 24–28 (2016)Google Scholar
  5. 5.
    Lan, R., Guo, S.C., Jia, S.Y.: Forensic image retrieval algorithm based on fusion. Comput. Eng. Des. 39(4), 1106–1110 (2018)Google Scholar
  6. 6.
    Liu, Y., Hu, D., Fan, J.L., et al.: Multi-feature fusion for crime scene investigation image retrieval. In: 2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA), pp. 1–7. IEEE (2017)Google Scholar
  7. 7.
    Liu, Y., Wang, F.-P., Hu, D., et al.: Multi-feature fusion with SVM classification for crime scene investigation image retrieval. In: IEEE International Conference on Signal & Image Processing, pp. 160–165. IEEE (2017)Google Scholar
  8. 8.
    Ezeobiejesi, J., Bhanu, B.: Latent fingerprint image quality assessment using deep learning. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 621–630. IEEE Computer Society (2018)Google Scholar
  9. 9.
    Vagac, M., Povinsky, M., Melichercik, M.: Detection of shoe sole features using DNN. In: 2017 IEEE 14th International Scientific Conference on Informatics, pp. 416–419. IEEE (2017)Google Scholar
  10. 10.
    Liu, Y., Peng, Y.N., et al.: A novel image retrieval algorithm based on transfer learning and fusion features. World Wide Web 22(3), 1313–1324 (2018)CrossRefGoogle Scholar
  11. 11.
    Kulkarni, P., Zepeda, J., Jurie, F., et al.: Hybrid multi-layer deep CNN/aggregator feature for image classification. In: IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 1379–1383. IEEE (2015)Google Scholar
  12. 12.
    Cai, D., Zhang, C., He, X.: Unsupervised feature selection for multi-cluster data. In: ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp. 333–342 (2010)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Qiannan Zhang
    • 1
    Email author
  • Ying Liu
    • 1
    • 2
  • Fuping Wang
    • 1
  • Jin Lu
    • 1
  • Daxiang Li
    • 1
  1. 1.School of Communication and Information EngineeringXi’an University of Posts and TelecommunicationsXi’anChina
  2. 2.Key Laboratory of Electronic Information Applications Technology for Scene InvestigationMinistry of Public SecurityXi’anChina

Personalised recommendations