Related Experiment Video
Updated: Oct 25, 2025
03:31
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
Published on: December 15, 2023
720
Attention-Based Multi-Scale Convolutional Neural Network (A+MCNN) for Multi-Class Classification in Road Images.
1Civil, Environmental, and Construction, Engineering Department, University of Central Florida, Orlando, FL 32816, USA.
Sensors (Basel, Switzerland)
|August 10, 2021
Summary
An attention-based multi-scale convolutional neural network (A+MCNN) enhances automated pavement distress recognition. This deep learning model improves road surface defect classification by incorporating contextual information.
Area of Science:
- Civil Engineering
- Computer Science
- Artificial Intelligence
Background:
- Automated pavement distress recognition is crucial for smart infrastructure assessment.
- Deep learning and computer vision have advanced pavement image analysis, but challenges remain due to defect variability.
- Effective recognition requires better incorporation of contextual information into deep networks.
Purpose of the Study:
- To introduce an attention-based multi-scale convolutional neural network (A+MCNN) for improved automated pavement distress classification.
- To enhance the encoding of contextual information and handle heterogeneous image contexts from different scales.
- To evaluate A+MCNN's performance against established deep classifiers in transportation applications.
Main Methods:
- Developed an attention-based multi-scale convolutional neural network (A+MCNN).
- Utilized multi-scale input tiles to encode contextual information.
- Employed a mid-fusion approach with an attention module for diverse image contexts.
- Trained and tested A+MCNN on distress (crack, crack seal, patch, pothole) and non-distress classes (joint, marker, manhole cover, curbing, shoulder) for asphalt and concrete pavements.
Main Results:
- A+MCNN demonstrated superior performance compared to four widely used deep classifiers and a generic CNN.
- The proposed model achieved an average F-score improvement of 1-26% over baseline models.
- Consistent outperformance was observed across various distress and non-distress pavement objects.
Conclusions:
- A+MCNN effectively improves automated pavement distress recognition by leveraging multi-scale context and attention mechanisms.
- The model's ability to integrate heterogeneous image contexts is key to its enhanced classification accuracy.
- This research offers valuable insights into classifier performance variations across different road objects, addressing a gap in existing literature.
Related Concept Videos
01:22Multi-input and Multi-variable systems
207
Cruise control systems in cars are designed as multi-input systems to maintain a driver's desired speed while compensating for external disturbances such as changes in terrain. The block diagram for a cruise control system typically includes two main inputs: the desired speed set by the driver and any external disturbances, such as the incline of the road. By adjusting the engine throttle, the system maintains the vehicle's speed as close to the desired value as possible.
In the absence...
In the absence...
207
01:20Parallel Processing
382
The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...
382