Remote Scene Image Scene Classification Based on Adaptive Segmentation and Dynamic Graph Convolution

Mapping Intimacies ◽

10.1109/igarss47720.2021.9553089 ◽

2021 ◽

Author(s):

Yuqun Yang ◽

Xu Tang ◽

Xiao Han ◽

Jingjing Ma ◽

Xiangrong Zhang ◽

...

Keyword(s):

Adaptive Segmentation ◽

Scene Classification ◽

Dynamic Graph ◽

Scene Image

Get full-text (via PubEx)

Scene Recognition Based on Recurrent Memorized Attention Network

Electronics ◽

10.3390/electronics9122038 ◽

2020 ◽

Vol 9 (12) ◽

pp. 2038

Author(s):

Xi Shao ◽

Xuan Zhang ◽

Guijin Tang ◽

Bingkun Bao

Keyword(s):

Recognition Accuracy ◽

State Of The Art ◽

Classification Performance ◽

Scene Recognition ◽

Experimental Results ◽

Scene Classification ◽

Attention Network ◽

Attention Model ◽

Scene Image ◽

Object Based

We propose a new end-to-end scene recognition framework, called a Recurrent Memorized Attention Network (RMAN) model, which performs object-based scene classification by recurrently locating and memorizing objects in the image. Based on the proposed framework, we introduce a multi-task mechanism that contiguously attends on the different essential objects in a scene image and recurrently performs memory fusion of the features of object focused by an attention model to improve the scene recognition accuracy. The experimental results show that the RMAN model has achieved better classification performance on the constructed dataset and two public scene datasets, surpassing state-of-the-art image scene recognition approaches.

Get full-text (via PubEx)

Multi-Feature Bio-Inspired Model for Scene Classification

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.303-306.1569 ◽

2013 ◽

Vol 303-306 ◽

pp. 1569-1572

Author(s):

De Kun Hu ◽

Jie Lin

Keyword(s):

Visual Cortex ◽

Image Classification ◽

Classification Accuracy ◽

Color Image ◽

High Accuracy ◽

Scene Classification ◽

Scene Image ◽

Image Dataset

A multi-feature bio-inspired model for scene image classification (MFBIM) is presented in this work; it extends the hierarchical feedforward model of the visual cortex. Firstly, each of three paths of classification uses each image property (i.e. shape, edge or color based features) independently. Then, BPNN assigns the category of an image based on the previous outputs. Experiments show that the model boosts the classification accuracy over the shape based model. Meanwhile, the proposed approach achieves a high accuracy comparable to other reported methods on publicly available color image dataset.

Get full-text (via PubEx)