Candev2018-aurora(单张图片的极光类型识别)

This project mainly want to recongize the type of aurora shown in picture automatically by machine!

Reference: https://github.com/Flaick/Candev2018-aurora

Dataset: https://open.canada.ca/data/en/dataset/d700c863-8622-4ec2-a4ee-a1c377880e2e

Why class2 and class6: http://tid.uio.no/plasma/oath/

This project implements on the Pytorch framework.

‘class2’ and ‘class6’ folders contain the dataset used for training two different models.

‘model’ folder contains the final model for the class2 classification.

class2new.py file contains the procedure of building a model and training the model with dataset.

convert.py file contains the procedure of making our own dataset.

Depends on the time and memory of GPU, we only train one model for ‘class2’ which is saved in the ‘model’ folder

The model is built on the Resnet-18

Structure of training data:
└───Candev
├───class2
│ ├───candev_data_2_0
│ └───candev_data_2_1
└───class6
├───candev_data_6_0
├───candev_data_6_1
├───candev_data_6_2
├───candev_data_6_3
├───candev_data_6_4
└───candev_data_6_5

Examples for each level:

class2:

candev_data_2_0:

图片来源于上述链接

candev_data_2_1:

图片来源于上述链接

class6:

candev_data_6_0:

图片来源于上述链接

candev_data_6_1:

图片来源于上述链接

candev_data_6_2:

图片来源于上述链接

candev_data_6_3:

图片来源于上述链接

candev_data_6_4:

图片来源于上述链接

candev_data_6_5:

图片来源于上述链接

After classifying the images we will combine the value of class2 and class6 to calculate the type of the picture!

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值