Scribd
Upload
8 views14 pages

Review2 2

The document presents a new deep learning method for object detection in autonomous vehicles under adverse weather conditions using a Faster Region Convolutional Neural Network. It addresses challenges posed by rain, fog, and snow by employing weather-aware data augmentation techniques to improve model generalization. The proposed system includes detailed training and testing phases for accurately localizing and classifying obstacles in degraded visual environments.

Uploaded by

gojodeku002
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
8 views14 pages

Review2 2

The document presents a new deep learning method for object detection in autonomous vehicles under adverse weather conditions using a Faster Region Convolutional Neural Network. It addresses challenges posed by rain, fog, and snow by employing weather-aware data augmentation techniques to improve model generalization. The proposed system includes detailed training and testing phases for accurately localizing and classifying obstacles in degraded visual environments.

Uploaded by

gojodeku002
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

Object Detection And Alert System For Complex Weather Road

Environment
ABSTRACT
Autonomous Vehicle (AV) technologies are faced with several challenges under
adverse weather conditions such as snow, fog, rain, sun glare, etc. Object detection
under adverse weather conditions is one of the most critical issues facing
autonomous driving. Such conditions are challenging for autonomous vehicles
because of the inability to track distinct visual features that may lead to road
accident. Propose a new deep learning method that can detect object or obstacles
from multiple bad weather degradations: rain, fog, snow and adherent raindrops
using a Faster Region Convolutional Neural Network architecture. The model
leverages a deep convolutional backbone with a Region Proposal Network (RPN) to
accurately localize objects in degraded visual environments. Additionally, weather-
aware data augmentation techniques are applied to enhance the model's
generalization capabilities across diverse weather scenarios.
MODEL FLOW – TRAINING PHASE

RCO Dataset
Preprocessing
Load

Resize

Noise Filter
Segmentation
Grey Trans.
FG Extraction
Binarization
BG Subtraction
FRCNN
Obstacle Detect Feature Extraction
Localize Size

Brach

Shape

Texture

Classification

Weather : Class

Obstacle : Class

Classified File
MODEL FLOW – TESTING PHASE
RCO Image Preprocessing

Load

Resize
Noise Filter
Segmentation
Grey Trans.
Binarization FG Extraction
BG Subtraction
Obstacle Detect Feature Extraction

Localize
Size
Branch
Shape
Texture

Load Classified Extracted OB


Prediction
File Threshold

Match

Obstacle No Obstacle

Obstacle
Weather

Alert
DATA FLOW DIAGRAM
Level 0
Level 1
Level 2
USECASE DIAGRAM
CLASS DIAGRAM
ACTIVITY DIAGRAM
SCREENSHOT
SCREENSHOT
SCREENSHOT
THANK YOU

You might also like