Deep Learning-Based Large-Scale Automatic Satellite Crosswalk Classification

@article{Berriel2017DeepLL,
  title={Deep Learning-Based Large-Scale Automatic Satellite Crosswalk Classification},
  author={Rodrigo Berriel and Andre Teixeira Lopes and Alberto F. de Souza and Thiago Oliveira-Santos},
  journal={IEEE Geoscience and Remote Sensing Letters},
  year={2017},
  volume={14},
  pages={1513-1517}
}
High-resolution satellite imagery has been increasingly used on remote sensing classification problems. One of the main factors is the availability of this kind of data. Despite the high availability, very little effort has been placed on the zebra crossing classification problem. In this letter, crowdsourcing systems are exploited in order to enable the automatic acquisition and annotation of a large-scale satellite imagery database for crosswalks related tasks. Then, this data set is used to… 

Figures and Tables from this paper

Satellite and Land Cover Image Classification using Deep Learning

TLDR
The problem of object and facility classification in satellite imagery is considered and the system is developed by using various facilities like Tensor Flow, XAMPP, FLASK and other various deep learning libraries.

Satellite Image Classification with Deep Learning: Survey

TLDR
The problem of object and facility recognition in satellite imagery is considered and a deep learning system consisting of an ensemble of convolutional neural networks and additional neural networks that integrate satellite metadata with image features is considered.

Satellite Imagery Classification with Deep Learning : A Survey

TLDR
The High-Resolution Satellite Imagery (HRSI) proved to be a suitable alternative to aerial photogrammetric data to provide a new data source for object detection and will be a better choice for automating such systems.

Heading Direction Estimation Using Deep Learning with Automatic Large-scale Data Acquisition

TLDR
The problem of estimating the heading direction that keeps the vehicle aligned with the road direction, which can be used in precise localization, road and lane keeping, lane departure warning, and others is addressed.

Deep Learning for Enrichment of Vector Spatial Databases

TLDR
This article proposes to use raster-based deep learning techniques to recognize highway interchanges using vector spatial data to study how to optimally convert vector data into small images suitable for state-of-the-art deep learning models.

Optical Remote Sensing Image Understanding with Weak Supervision: Concepts, Methods, and Perspectives

In recent years, supervised learning has been widely used in various tasks of optical remote sensing image (RSI) understanding, including RSI classification, pixel-wise segmentation, change

Crowdsourcing in Remote Sensing: A Review of Applications and Future Directions

TLDR
The data obtained with remote sensing sensors are processed to fuel studies related to the Earth's resources and environment, and the analysis and interpretation of most satellite data have not yet been fully automated, and human intervention is still necessary at certain stages.

Cross-Domain Car Detection Using Unsupervised Image-to-Image Translation: From Day to Night

TLDR
A model based on Generative Adversarial Networks (GANs) is explored to enable the generation of an artificial dataset with its respective annotations that is used to train a car detector model with annotated data from a source domain without requiring the image annotations of the target domain.

References

SHOWING 1-10 OF 13 REFERENCES

Crosswalk Localization from Low Resolution Satellite Images to Assist Visually Impaired People

TLDR
Experimental results indicate that the proposed model for crosswalk detection and localization by using satellite images captured from Google Maps works well in low resolution images, effectively detecting and localizing crosswalks in simulated scenarios.

Zebra Crossing Detection from Aerial Imagery Across Countries

TLDR
The usefulness of the proposed approach to detect zebra crossings in aerial imagery is shown, and the system automatically learns an appearance model from available geospatial data for an examined region without requiring any additional data for that specific region.

Very Deep Convolutional Networks for Large-Scale Image Recognition

TLDR
This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.

ImageNet classification with deep convolutional neural networks

TLDR
A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.

ImageNet Large Scale Visual Recognition Challenge

TLDR
The creation of this benchmark dataset and the advances in object recognition that have been possible as a result are described, and the state-of-the-art computer vision accuracy with human accuracy is compared.

Urban Road Network Extraction Based on Zebra Crossing Detection from a Very High Resolution RGB Aerial Image and DSM Data

TLDR
This work uses a circle mask template matching and Speeded Up Robust Features (SURF) method in order to detect and evaluate the zebra crossing location in an RGB aerial image and proposes a road extraction based on zebra crossings detection.

Going deeper with convolutions

We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition

Zebra Crossing Spotter: Automatic Population of Spatial Databases for Increased Safety of Blind Travelers

TLDR
A computer vision-based technique that mines existing spatial image databases for discovery of zebra crosswalks in urban settings and is complemented by a final crowdsourcing validation stage for increased accuracy.

Detecting and locating crosswalks using a camera phone

TLDR
The novel ldquoCrosswatchldquo system, which uses computer vision to provide information about the location and orientation of crosswalks to a blind or visually impaired pedestrian holding a camera cell phone, is described.

On visual crosswalk detection for driver assistance systems

TLDR
A new crosswalk detection strategy is presented that is narrowed down to driver assistance systems to guarantee a reliable detection result and to benefit from the properties of a vehicle mounted camera.