Contrail Recognition with Convolutional Neural Network and ...
Plant Recognition using Convolutional Neural Networks
Transcript of Plant Recognition using Convolutional Neural Networks
![Page 1: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/1.jpg)
Plant Recognition using Convolutional Neural Networks Qiming Chen, Zhiyuan Guo
1
![Page 2: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/2.jpg)
Overview
Problem Statement
Related Works
Analysis of Dataset
Implementation
Future works
2
![Page 3: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/3.jpg)
Problem Statement
From an organ to the entire
3
![Page 4: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/4.jpg)
Problem Statement
4
![Page 5: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/5.jpg)
Analysis of Dataset
2013 2014 2015
250 500 1000
26077 60000 113205
Year:
Species:
Pictures:
5
![Page 6: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/6.jpg)
Analysis of Dataset
5476 25972
28225 7720
8130 16235 6
![Page 7: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/7.jpg)
Related Works
7
![Page 8: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/8.jpg)
Related Works
IBM Team
Pl@antNet Team
FINKI Team
8
![Page 9: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/9.jpg)
IBM Team
SIFT GMM Fisher Encoding SVM
CNN RUN1:
RUN2:
RUN3: Combine RUN1&2
RUN4: Add ROI on RUN3 9
![Page 10: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/10.jpg)
Pl@ant Team
10
![Page 11: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/11.jpg)
FINKI Team
11
![Page 12: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/12.jpg)
Implementation
Corp the image:
300*400 on average —> 128*128
60 GB —> 18 GB
12
![Page 13: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/13.jpg)
Hyper parameters Network Shape:
Input: (128, 128) * 3
Conv: (64, 64) * 48
Conv: (31, 31) * 64
Conv: (16, 16) * 96
Conv: (16, 16) * 64
Full-con: 1024 (num_pieces: 5)
Softmax Output: 1000 13
![Page 14: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/14.jpg)
Hyper parameters Kernel Shape: 9*9, 5*5, 3*3, 3*3
Max Pooling: 2*2, 4*4, 2*2, no pooling
Learning Rate: 0.005
Linear Decay: 0.01 over 250 Epoch
Learning rate for conv layers: scale 0.05
Momentum: 0.5
L2 norm: Weight 0.00005 on each layer 14
![Page 15: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/15.jpg)
Our CNN
15
![Page 16: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/16.jpg)
IBM CNN
16
![Page 17: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/17.jpg)
Best validation precision: 19.31%
17
![Page 18: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/18.jpg)
How is 19% ?
18
![Page 19: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/19.jpg)
How is 19% ? The metrics of the competition:
r = rank of the ground truth in the top n predictions
Score = Average ( 1/r )
19
![Page 20: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/20.jpg)
Some problems “Deep Learning is easy! Just buy chips and tune parameters!”
It’s hard to make it work.
I ain’t got no chips !!!
Memory certainly not big enough for the data.
If we don’t crop the images, the hard drive won’t fit.
1h/epoch with 4 conv layers.
GPU memory not enough.
20
![Page 21: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/21.jpg)
Some problems Is the data enough?
60 pics / class
10 pics / (class, view)
21
![Page 22: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/22.jpg)
Future work
Drop out
Complete the testing run and compete in the contest
22
![Page 23: Plant Recognition using Convolutional Neural Networks](https://reader030.fdocuments.in/reader030/viewer/2022012706/61a7b028dcb3b309596a3d0e/html5/thumbnails/23.jpg)
Thanks
23