A Hybrid Geometric Spatial Image Representation for scene classification

Abstract

The recent development in the technology has increased the complexity of image contents and demand for image classification becomes more imperative. Digital images play a vital role in many applied domains such as remote sensing, scene analysis, medical care, textile industry and crime investigation. Feature extraction and image representation is considered as an important step in scene analysis as it affects the image classification performance. Automatic classification of images is an open research problem for image analysis and pattern recognition applications. The Bag-of-Features (BoF) model is commonly used to solve image classification, object recognition and other computer vision-based problems. In BoF model, the final feature vector representation of an image contains no information about the co-occurrence of features in the 2D image space. This is considered as a limitation, as the spatial arrangement among visual words in image space contains the information that is beneficial for image representation and learning of classification model. To deal with this, researchers have proposed different image representations. Among these, the division of image-space into different geometric sub-regions for the extraction of histogram for BoF model is considered as a notable contribution for the extraction of spatial clues. Keeping this in view, we aim to explore a Hybrid Geometric Spatial Image Representation (HGSIR) that is based on the combination of histograms computed over the rectangular, triangular and circular regions of the image. Five standard image datasets are used to evaluate the performance of the proposed research. The quantitative analysis demonstrates that the proposed research outperforms the state-of-art research in terms of classification accuracy.

Fig 1. Images taken from the different classes of MSRC-v2 image database [18].

Fig 2. Fig (a) shows the approach based on geometric relationships among visual words [14] (b) SPM approach based on histograms of geometric sub-regions (rectangular) [10].

Fig 3. The block diagram of proposed research based on HGSIR.

Fig 4. The photo gallery of images representing each class of 15-scene image dataset.

Fig 5. The photo gallery of images representing each class of UCM dataset.

Fig 6. The photo gallery of images selected from the Caltech-101 dataset.

Fig 7. The photo gallery of images selected from the RSSCN7 image dataset.

Fig 8. The photo gallery of images selected from the MSRC-v2 image dataset.

Fig 9. The mean classification accuracy comparison while using different sizes of visual vocabulary for 15-scene image dataset.

Fig 10. The confusion matrix representing the computed classification accuracy % for the proposed research while using 15-scene image dataset.

Fig 11. Class-wise comparison between LGF [38] and HGSIR for the 15-scene image dataset.

Fig 12. The mean classification accuracy comparison while using different sizes of visual vocabulary for UCM image dataset.

Fig 13. The confusion matrix representing the computed classification accuracy % for the proposed research while using UCM image dataset.

Fig 14. Class-wise comparison between LGF [38] and HGSIR for UCM image dataset.

Fig 15. The mean classification accuracy comparison while using different sizes of visual vocabulary for Caltech-101 image dataset.

Fig 16. The mean classification accuracy comparison while using different sizes of visual vocabulary for RSSCN7 image dataset.

Fig 17. The confusion matrix representing the computed classification accuracy % for the proposed research while using RSSCN7 dataset.

Fig 18. The confusion matrix representing the computed classification accuracy % for the proposed research while using MSRC-v2 image dataset.