Scene Classification with a Biologically Inspired Method
We present a biologically motivated method for scene image classification. The core of the method is to use shape based image property that is provided by a hierarchical feedforward model of the visual cortex [18]. Edge based and color based image properties are additionally used to improve the accu...
| Main Author: | |
|---|---|
| Other Authors: | |
| Published: |
2009
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/45516 |
| _version_ | 1811072837030510592 |
|---|---|
| author | Terashima, Yoshito |
| author2 | Tomaso Poggio |
| author_facet | Tomaso Poggio Terashima, Yoshito |
| author_sort | Terashima, Yoshito |
| collection | MIT |
| description | We present a biologically motivated method for scene image classification. The core of the method is to use shape based image property that is provided by a hierarchical feedforward model of the visual cortex [18]. Edge based and color based image properties are additionally used to improve the accuracy. The method consists of two stages of image analysis. In the first stage, each of three paths of classification uses each image property (i.e. shape, edge or color based features) independently. In the second stage, a single classifier assigns the category of an image based on the probability distributions of the first stage classifier outputs. Experiments show that the method boosts the classification accuracy over the shape based model. We demonstrate that this method achieves a high accuracy comparable to other reported methods on publicly available color image dataset. |
| first_indexed | 2024-09-23T09:16:08Z |
| id | mit-1721.1/45516 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T09:16:08Z |
| publishDate | 2009 |
| record_format | dspace |
| spelling | mit-1721.1/455162019-04-11T00:32:20Z Scene Classification with a Biologically Inspired Method Terashima, Yoshito Tomaso Poggio Center for Biological and Computational Learning (CBCL) image classification vision We present a biologically motivated method for scene image classification. The core of the method is to use shape based image property that is provided by a hierarchical feedforward model of the visual cortex [18]. Edge based and color based image properties are additionally used to improve the accuracy. The method consists of two stages of image analysis. In the first stage, each of three paths of classification uses each image property (i.e. shape, edge or color based features) independently. In the second stage, a single classifier assigns the category of an image based on the probability distributions of the first stage classifier outputs. Experiments show that the method boosts the classification accuracy over the shape based model. We demonstrate that this method achieves a high accuracy comparable to other reported methods on publicly available color image dataset. 2009-05-11T17:30:10Z 2009-05-11T17:30:10Z 2009-05-10 http://hdl.handle.net/1721.1/45516 CBCL-277 MIT-CSAIL-TR-2009-020 8 p. application/pdf application/postscript |
| spellingShingle | image classification vision Terashima, Yoshito Scene Classification with a Biologically Inspired Method |
| title | Scene Classification with a Biologically Inspired Method |
| title_full | Scene Classification with a Biologically Inspired Method |
| title_fullStr | Scene Classification with a Biologically Inspired Method |
| title_full_unstemmed | Scene Classification with a Biologically Inspired Method |
| title_short | Scene Classification with a Biologically Inspired Method |
| title_sort | scene classification with a biologically inspired method |
| topic | image classification vision |
| url | http://hdl.handle.net/1721.1/45516 |
| work_keys_str_mv | AT terashimayoshito sceneclassificationwithabiologicallyinspiredmethod |