Indian Traffic Sign Detection Benchmark Dataset in YOLO Format

0
0 ratings - Please login to submit your rating.

Abstract 

This is the collection of Indian Traffic Sign Detection Dataset. This can be used maily on Traffic Sign detection projects using YOLO. Dataset is in YOLO format. There are 1264 total images in this dataset fully annotated using Labelimg tool. Some augmented datas using techniques like blurring, mosaic etc.. are also present. The dataset has images in 3 different types of traffic signs in India. Dataset is annotated only as one class-Traffic Sign. The dataset produced a Mean Average Precision(mAP) of 99.28 % using yolov4 and 99.20 % using yolov3 at 50% Intersection over Union(IoU) threshold. The most commonly used German Traffic sign Dataset is having only 900 images. Also increase in dataset and reducing the no of classes helped reduce training time.

Instructions: 

Indian Traffic Sign Detection Dataset is in YOLO format. There are 1264 total images in this dataset fully annotated using Labelimg tool. Some augmented datas using techniques like blurring, mosaic etc.. are also present. The dataset has images in 3 different types of traffic signs in India. Dataset is annotated only as one class-Traffic Sign. The dataset produced a Mean Average Precision(mAP) of 99.28 % using yolov4 and 99.20 % using yolov3 at 50% Intersection over Union(IoU) threshold. The most commonly used German Traffic sign Dataset is having only 900 images. Also increase in dataset and reducing the no of classes helped reduce training time.

Comments

I can't access it

Submitted by Adriana Miu on Sat, 02/12/2022 - 09:09

I didn't found any files to access

Submitted by Kavya Gangula on Tue, 02/15/2022 - 09:01

mb

Submitted by ameer massalha on Sun, 02/27/2022 - 06:22

I didn't find the files to access.

Submitted by Nikita Chadha on Sun, 10/02/2022 - 10:23

Can I get this dataset?

Submitted by HANEESH NAGA KR... on Wed, 07/26/2023 - 05:35