Digital Image Processing: Homework #5

EE 569 Digital Image Processing: Homework #5

EE 569: Homework #5

General Instructions:
1. Read Homework Guidelines for the information about homework programming, write-up and
submission. If you make any assumptions about a problem, please clearly state them in your report.
2. You are required to use PYTHON in this assignment. It is recommended to use interface tool
PYTORCH. KERAS is an alternative choice if you feel more comfortable with it, which is built upon
TENSORFLOW. We only provide sample tutorial using PYTORCH.
3. DO NOT copy codes from online sources e.g. Github.
4. You need to understand the USC policy on academic integrity and penalties for cheating and
plagiarism. These rules will be strictly enforced.
Problem 1: CNN Training on LeNet-5 (100%)
In this problem, you will learn to train a simple convolutional neural network (CNN) called the LeNet-5,
introduced by LeCun et al. [1], and apply it to three datasets MNIST [2], Fashion-MNIST [3] and
CIFAR-10 [4].
LeNet-5 is designed for handwritten and machine-printed character recognition. Its architecture is shown
in Fig. 1. This network has two conv layers, and three fc layers. Each conv layer is followed by a max
pooling layer. Both conv layers accept an input receptive field of spatial size 5x5. The filter numbers of
the first and the second conv layers are 6 and 16 respectively. The stride parameter is 1 and no padding is
used. The two max pooling layers take an input window size of 2x2, reduce the window size to 1x1 by
choosing the maximum value of the four responses. The first two fc layers have 120 and 84 filters,
respectively. The last fc layer, the output layer, has size of 10 to match the number of object classes in the
dataset. Use the popular ReLU activation function [5] for all conv and all fc layers except for the output
layer, which uses softmax [6] to compute the probabilities.
Figure 1: A CNN architecture derived from LeNet-5
EE 569 Digital Image Processing: Homework #5
Professor C.-C. Jay Kuo Page 2 of 3
The following table shows statistics for different datasets:
Image type Image size # Class # training
images
# testing
images
MNIST Gray 28*28 10 60,000 10,000
FashionMNIST
Gray 28*28 10 60,000 10,000
CIFAR-10 Color 32*32 10 50,000 10,000
(a) CNN Architecture (Basic: 20%)
Explain the architecture and operational mechanism of convolutional neural networks by performing the
following tasks.
1. Describe CNN components in your own words: 1) the fully connected layer, 2) the convolutional
layer, 3) the max pooling layer, 4) the activation function, and 5) the softmax function. What are
the functions of these components?
2. What is the over-fitting issue in model learning? Explain any technique that has been used in CNN
training to avoid the over-fitting.
3. Explain the difference among different activation functions including ReLU, LeakyReLU and
ELU.
4. Read official documents of different loss functions including L1Loss, MSELoss and BCELoss.
List applications where those losses are used, and state why do you think they are used in those
specific cases?
Show your understanding as much as possible in your own words in your report.
(b) Compare classification performance on different datasets (Basic: 50%)
Train the CNN given in Fig. 1 using the training images of MNIST, then test the trained network on the
testing images of MNIST. Compute and draw the accuracy performance curves (epoch-accuracy plot) on
training and test datasets on the same figure. You can adopt proper preprocessing techniques and the
random network initialization to make your training work easy.
1. Plot the performance curves under 5 different yet representative initial parameter settings
(initialization of filter weights, learning rate, decay and etc.). Discuss your observations and the
effect of different settings.
2. Find the best parameter setting to achieve the highest accuracy on the test set. Then, plot the
performance curves for the test set and the training set under this setting. Your testing accuracy
should be no less than 99%.
3. Repeat 1 and 2 for Fashion-MNIST. Your best testing accuracy should be no less than 90%.
4. Repeat 1 and 2 for CIFAR-10. Your best testing accuracy should be no less than 65%.
5. Compare your best performances on three datasets. How do they differ and why do you think there
is such difference?
EE 569 Digital Image Processing: Homework #5
Professor C.-C. Jay Kuo Page 3 of 3
(c) Apply trained network to negative images (Advanced: 30%)
You may achieve good recognition performance on the MNIST dataset in Problem 1(b). Do you think the
LeNet-5 understands the handwritten digits as well as human beings? One test is to provide a negative of
each test image as shown in Fig. 3, where the value of the negative image at pixel (x,y), denoted by r(x,y),
is computed via r(x,y)=255-p(x,y), where p(x,y) is the value of the original image at the same location.
Humans have no difficulty in recognizing digits of both types. How about the LeNet-5?
Figure 2: Sample images from original MNIST dataset
Figure 3: Sample images from the negatives of the MNIST dataset
1. Describe how you can get negatives of the testing set. Implement your idea, then use statistics and
sample images to show that you correctly reverse the intensity.
2. Report the accuracy on the negative test images using the LeNet-5 trained in part b). Discuss your
result.
3. Design and train a new network that can recognize both original and negative images from the
MNIST test dataset. Test your proposed network, report the accuracy and make discussion.
References
[1] LeCun, Yann, et al. "Gradient-based learning applied to document recognition." Proceedings of the
IEEE 86.11 (1998): 2278-2324
[2] http://yann.lecun.com/exdb/mnist/
[3] https://github.com/zalandoresearch/fashion-mnist
[4] https://www.cs.toronto.edu/~kriz/cifar.html
[5] ReLU https://en.wikipedia.org/wiki/Rectifier_(neural_networks).
[6] Softmax https://en.wikipedia.org/wiki/Softmax_function