Deep Learning_Autonomous driving application - Car detection_自動駕駛代碼實現

Autonomous driving - Car detection

Welcome to your week 3 programming assignment. You will learn about object detection using the very powerful YOLO model. Many of the ideas in this notebook are described in the two YOLO papers: Redmon et al., 2016 (https://arxiv.org/abs/1506.02640) and Redmon and Farhadi, 2016 (https://arxiv.org/abs/1612.08242).

You will learn to:

• Use object detection on a car detection dataset
• Deal with bounding boxes

Run the following cell to load the packages and dependencies that are going to be useful for your journey!

import argparse
import os
import matplotlib.pyplot as plt
from matplotlib.pyplot import imshow
import scipy.io
import scipy.misc
import numpy as np
import pandas as pd
import PIL
import tensorflow as tf
from keras import backend as K
from keras.layers import Input, Lambda, Conv2D
from keras.models import load_model, Model
from yolo_utils import read_classes, read_anchors, generate_colors, preprocess_image, draw_boxes, scale_boxes
from yad2k.models.keras_yolo import yolo_head, yolo_boxes_to_corners, preprocess_true_boxes, yolo_loss, yolo_body

%matplotlib inline

1 - Problem Statement

You are working on a self-driving car. As a critical component of this project, you'd like to first build a car detection system. To collect data, you've mounted a camera to the hood (meaning the front) of the car, which takes pictures of the road ahead every few seconds while you drive around.

2 - YOLO

YOLO ("you only look once") is a popular algoritm because it achieves high accuracy while also being able to run in real-time. This algorithm "only looks once" at the image in the sense that it requires only one forward propagation pass through the network to make predictions. After non-max suppression, it then outputs recognized objects together with the bounding boxes.

2.1 - Model details

First things to know:

• The input is a batch of images of shape (m, 608, 608, 3)
• The output is a list of bounding boxes along with the recognized classes. Each bounding box is represented by 6 numbers (pc,bx,by,bh,bw,c)(pc,bx,by,bh,bw,c) as explained above. If you expand cc into an 80-dimensional vector, each bounding box is then represented by 85 numbers.

We will use 5 anchor boxes. So you can think of the YOLO architecture as the following: IMAGE (m, 608, 608, 3) -> DEEP CNN -> ENCODING (m, 19, 19, 5, 85).

Lets look in greater detail at what this encoding represents.

2.2 - Filtering with a threshold on class scores

You are going to apply a first filter by thresholding. You would like to get rid of any box for which the class "score" is less than a chosen threshold.

The model gives you a total of 19x19x5x85 numbers, with each box described by 85 numbers. It'll be convenient to rearrange the (19,19,5,85) (or (19,19,425)) dimensional tensor into the following variables:

• box_confidence: tensor of shape (19×19,5,1)(19×19,5,1) containing pcpc (confidence probability that there's some object) for each of the 5 boxes predicted in each of the 19x19 cells.
• boxes: tensor of shape (19×19,5,4)(19×19,5,4) containing (bx,by,bh,bw)(bx,by,bh,bw) for each of the 5 boxes per cell.
• box_class_probs: tensor of shape (19×19,5,80)(19×19,5,80) containing the detection probabilities (c1,c2,...c80)(c1,c2,...c80) for each of the 80 classes for each of the 5 boxes per cell.

Exercise: Implement yolo_filter_boxes().

1. Compute box scores by doing the elementwise product as described in Figure 4. The following code may help you choose the right operator:

   a = np.random.randn(19*19, 5, 1)
   b = np.random.randn(19*19, 5, 80)
   c = a * b # shape of c will be (19*19, 5, 80)
   

2. For each box, find:
   • the index of the class with the maximum box score (Hint) (Be careful with what axis you choose; consider using axis=-1)
   • the corresponding box score (Hint) (Be careful with what axis you choose; consider using axis=-1)
3. Create a mask by using a threshold. As a reminder: ([0.9, 0.3, 0.4, 0.5, 0.1] < 0.4) returns: [False, True, False, False, True]. The mask should be True for the boxes you want to keep.
4. Use TensorFlow to apply the mask to box_class_scores, boxes and box_classes to filter out the boxes we don't want. You should be left with just the subset of boxes you want to keep. (Hint)

Reminder: to call a Keras function, you should use K.function(...).

# GRADED FUNCTION: yolo_filter_boxes

def yolo_filter_boxes(box_confidence, boxes, box_class_probs, threshold = .6):
    """Filters YOLO boxes by thresholding on object and class confidence.
    
    Arguments:
    box_confidence -- tensor of shape (19, 19, 5, 1)
    boxes -- tensor of shape (19, 19, 5, 4)
    box_class_probs -- tensor of shape (19, 19, 5, 80)
    threshold -- real value, if [ highest class probability score < threshold], then get rid of the corresponding box
    
    Returns:
    scores -- tensor of shape (None,), containing the class probability score for selected boxes
    boxes -- tensor of shape (None, 4), containing (b_x, b_y, b_h, b_w) coordinates of selected boxes
    classes -- tensor of shape (None,), containing the index of the class detected by the selected boxes
    
    Note: "None" is here because you don't know the exact number of selected boxes, as it depends on the threshold. 
    For example, the actual output size of scores would be (10,) if there are 10 boxes.
    """
    
    # Step 1: Compute box scores
    ### START CODE HERE ### (≈ 1 line)
    box_scores = box_confidence*box_class_probs
    ### END CODE HERE ###
    
    # Step 2: Find the box_classes thanks to the max box_scores, keep track of the corresponding score
    ### START CODE HERE ### (≈ 2 lines)
    box_classes = K.argmax(box_scores,axis=-1) 
    box_class_scores = K.max(box_scores, axis=-1)
    ### END CODE HERE ###
    
    # Step 3: Create a filtering mask based on "box_class_scores" by using "threshold". The mask should have the
    # same dimension as box_class_scores, and be True for the boxes you want to keep (with probability >= threshold)
    ### START CODE HERE ### (≈ 1 line)
    filtering_mask = box_class_scores>=threshold 
    ### END CODE HERE ###
    
    # Step 4: Apply the mask to scores, boxes and classes
    ### START CODE HERE ### (≈ 3 lines)
    scores = tf.boolean_mask(box_class_scores,filtering_mask)
    boxes = tf.boolean_mask(boxes,filtering_mask)
    classes = tf.boolean_mask(box_classes,filtering_mask)
    ### END CODE HERE ###
    
    return scores, boxes, classes

tf.reset_default_graph()

with tf.Session() as test:
    tf.set_random_seed(1)
    a_C = tf.random_normal([1, 4, 4, 3], mean=1, stddev=4)
    a_G = tf.random_normal([1, 4, 4, 3], mean=1, stddev=4)
    J_content = compute_content_cost(a_C, a_G)
    print("J_content = " + str(J_content.eval()))

Expected Output:

J_content

6.76559

What you should remember:

The content cost takes a hidden layer activation of the neural network, and measures how different a(C)a(C) and a(G)a(G) are.When we minimize the content cost later, this will help make sure GG has similar content as CC.

3.2.1 - Style matrix

The style matrix is also called a "Gram matrix." In linear algebra, the Gram matrix G of a set of vectors (v1,…,vn)(v1,…,vn) is the matrix of dot products, whose entries are Gij=vTivj=np.dot(vi,vj)Gij=viTvj=np.dot(vi,vj). In other words, GijGij compares how similar vivi is to vjvj: If they are highly similar, you would expect them to have a large dot product, and thus for GijGij to be large.

Note that there is an unfortunate collision in the variable names used here. We are following common terminology used in the literature, but GG is used to denote the Style matrix (or Gram matrix) as well as to denote the generated image GG. We will try to make sure which GG we are referring to is always clear from the context.

In NST, you can compute the Style matrix by multiplying the "unrolled" filter matrix with their transpose:

The result is a matrix of dimension (nC,nC)(nC,nC) where nCnC is the number of filters. The value GijGij measures how similar the activations of filter ii are to the activations of filter jj.

One important part of the gram matrix is that the diagonal elements such as GiiGii also measures how active filter ii is. For example, suppose filter ii is detecting vertical textures in the image. Then GiiGii measures how common vertical textures are in the image as a whole: If GiiGii is large, this means that the image has a lot of vertical texture.

By capturing the prevalence of different types of features (GiiGii), as well as how much different features occur together (GijGij), the Style matrix GG measures the style of an image.

Exercise: Using TensorFlow, implement a function that computes the Gram matrix of a matrix A. The formula is: The gram matrix of A is GA=AATGA=AAT. If you are stuck, take a look at Hint 1 and Hint 2.

# GRADED FUNCTION: gram_matrix

def gram_matrix(A):
    """
    Argument:
    A -- matrix of shape (n_C, n_H*n_W)
    
    Returns:
    GA -- Gram matrix of A, of shape (n_C, n_C)
    """
    
    ### START CODE HERE ### (≈1 line)
    GA = tf.matmul(A,tf.transpose(A))
    ### END CODE HERE ###
    
    return GA

tf.reset_default_graph()

with tf.Session() as test:
    tf.set_random_seed(1)
    A = tf.random_normal([3, 2*1], mean=1, stddev=4)
    GA = gram_matrix(A)
    
    print("GA = " + str(GA.eval()))

3.2.2 - Style cost

After generating the Style matrix (Gram matrix), your goal will be to minimize the distance between the Gram matrix of the "style" image S and that of the "generated" image G. For now, we are using only a single hidden layer a[l]a[l], and the corresponding style cost for this layer is defined as:

J[l]style(S,G)=14×nC2×(nH×nW)2∑i=1nC∑j=1nC(G(S)ij−G(G)ij)2(2)(2)Jstyle[l](S,G)=14×nC2×(nH×nW)2∑i=1nC∑j=1nC(Gij(S)−Gij(G))2

where G(S)G(S) and G(G)G(G) are respectively the Gram matrices of the "style" image and the "generated" image, computed using the hidden layer activations for a particular hidden layer in the network.

Exercise: Compute the style cost for a single layer.

Instructions: The 3 steps to implement this function are:

1. Retrieve dimensions from the hidden layer activations a_G:
   • To retrieve dimensions from a tensor X, use: X.get_shape().as_list()
2. Unroll the hidden layer activations a_S and a_G into 2D matrices, as explained in the picture above.
   • You may find Hint1 and Hint2 useful.
3. Compute the Style matrix of the images S and G. (Use the function you had previously written.)
4. Compute the Style cost:
   • You may find Hint3, Hint4 and Hint5 useful.

# GRADED FUNCTION: compute_layer_style_cost

def compute_layer_style_cost(a_S, a_G):
    """
    Arguments:
    a_S -- tensor of dimension (1, n_H, n_W, n_C), hidden layer activations representing style of the image S 
    a_G -- tensor of dimension (1, n_H, n_W, n_C), hidden layer activations representing style of the image G
    
    Returns: 
    J_style_layer -- tensor representing a scalar value, style cost defined above by equation (2)
    """
    
    ### START CODE HERE ###
    # Retrieve dimensions from a_G (≈1 line)
    m, n_H, n_W, n_C = a_G.get_shape().as_list()
    
    # Reshape the images to have them of shape (n_C, n_H*n_W) (≈2 lines)
    a_S = tf.transpose(tf.reshape(a_S,shape=[n_H*n_W,n_C]))
    a_G = tf.transpose(tf.reshape(a_G,shape=[n_H*n_W,n_C]))

    # Computing gram_matrices for both images S and G (≈2 lines)
    GS = gram_matrix(a_S)
    GG = gram_matrix(a_G)

    # Computing the loss (≈1 line)
    J_style_layer = tf.reduce_sum(tf.square(tf.subtract(GS,GG)))/(4*(n_H*n_W)**2*(n_C)**2)
    
    ### END CODE HERE ###
    
    return J_style_layer

tf.reset_default_graph()

with tf.Session() as test:
    tf.set_random_seed(1)
    a_S = tf.random_normal([1, 4, 4, 3], mean=1, stddev=4)
    a_G = tf.random_normal([1, 4, 4, 3], mean=1, stddev=4)
    J_style_layer = compute_layer_style_cost(a_S, a_G)
    
    print("J_style_layer = " + str(J_style_layer.eval()))

Expected Output:

J_style_layer

9.19028

3.2.3 Style Weights

So far you have captured the style from only one layer. We'll get better results if we "merge" style costs from several different layers. After completing this exercise, feel free to come back and experiment with different weights to see how it changes the generated image GG. But for now, this is a pretty reasonable default:

In [ ]:

You can combine the style costs for different layers as follows:

Jstyle(S,G)=∑lλ[l]J[l]style(S,G)Jstyle(S,G)=∑lλ[l]Jstyle[l](S,G)

where the values for λ[l]λ[l] are given in STYLE_LAYERS.

We've implemented a compute_style_cost(...) function. It simply calls your compute_layer_style_cost(...) several times, and weights their results using the values in STYLE_LAYERS. Read over it to make sure you understand what it's doing.

def compute_style_cost(model, STYLE_LAYERS):
    """
    Computes the overall style cost from several chosen layers
    
    Arguments:
    model -- our tensorflow model
    STYLE_LAYERS -- A python list containing:
                        - the names of the layers we would like to extract style from
                        - a coefficient for each of them
    
    Returns: 
    J_style -- tensor representing a scalar value, style cost defined above by equation (2)
    """
    
    # initialize the overall style cost
    J_style = 0

    for layer_name, coeff in STYLE_LAYERS:

        # Select the output tensor of the currently selected layer
        out = model[layer_name]

        # Set a_S to be the hidden layer activation from the layer we have selected, by running the session on out
        a_S = sess.run(out)

        # Set a_G to be the hidden layer activation from same layer. Here, a_G references model[layer_name] 
        # and isn't evaluated yet. Later in the code, we'll assign the image G as the model input, so that
        # when we run the session, this will be the activations drawn from the appropriate layer, with G as input.
        a_G = out
        
        # Compute style_cost for the current layer
        J_style_layer = compute_layer_style_cost(a_S, a_G)

        # Add coeff * J_style_layer of this layer to overall style cost
        J_style += coeff * J_style_layer

    return J_style

Note: In the inner-loop of the for-loop above, a_G is a tensor and hasn't been evaluated yet. It will be evaluated and updated at each iteration when we run the TensorFlow graph in model_nn() below.

What you should remember:

The style of an image can be represented using the Gram matrix of a hidden layer's activations. However, we get even better results combining this representation from multiple different layers. This is in contrast to the content representation, where usually using just a single hidden layer is sufficient.Minimizing the style cost will cause the image GG to follow the style of the image SS.

3.3 - Defining the total cost to optimize

Finally, let's create a cost function that minimizes both the style and the content cost. The formula is:

J(G)=αJcontent(C,G)+βJstyle(S,G)J(G)=αJcontent(C,G)+βJstyle(S,G)

Exercise: Implement the total cost function which includes both the content cost and the style cost.

# GRADED FUNCTION: total_cost

def total_cost(J_content, J_style, alpha = 10, beta = 40):
    """
    Computes the total cost function
    
    Arguments:
    J_content -- content cost coded above
    J_style -- style cost coded above
    alpha -- hyperparameter weighting the importance of the content cost
    beta -- hyperparameter weighting the importance of the style cost
    
    Returns:
    J -- total cost as defined by the formula above.
    """
    
    ### START CODE HERE ### (≈1 line)
    J = alpha*J_content+beta*J_style
    ### END CODE HERE ###
    
    return J

tf.reset_default_graph()

with tf.Session() as test:
    np.random.seed(3)
    J_content = np.random.randn()    
    J_style = np.random.randn()
    J = total_cost(J_content, J_style)
    print("J = " + str(J))

4 - Solving the optimization problem

Finally, let's put everything together to implement Neural Style Transfer!

Here's what the program will have to do:

1. Create an Interactive Session
2. Load the content image
3. Load the style image
4. Randomly initialize the image to be generated
5. Load the VGG16 model
6. Build the TensorFlow graph:
   • Run the content image through the VGG16 model and compute the content cost
   • Run the style image through the VGG16 model and compute the style cost
   • Compute the total cost
   • Define the optimizer and the learning rate
7. Initialize the TensorFlow graph and run it for a large number of iterations, updating the generated image at every step.

Lets go through the individual steps in detail.

You've previously implemented the overall cost J(G)J(G). We'll now set up TensorFlow to optimize this with respect to GG. To do so, your program has to reset the graph and use an "Interactive Session". Unlike a regular session, the "Interactive Session" installs itself as the default session to build a graph. This allows you to run variables without constantly needing to refer to the session object, which simplifies the code.

Lets start the interactive session.

# Reset the graph
tf.reset_default_graph()

# Start interactive session
sess = tf.InteractiveSession()

Let's load, reshape, and normalize our "content" image (the Louvre museum picture):

content_image = scipy.misc.imread("images/louvre_small.jpg")
content_image = reshape_and_normalize_image(content_image)

Now, we initialize the "generated" image as a noisy image created from the content_image. By initializing the pixels of the generated image to be mostly noise but still slightly correlated with the content image, this will help the content of the "generated" image more rapidly match the content of the "content" image. (Feel free to look in nst_utils.py to see the details of generate_noise_image(...); to do so, click "File-->Open..." at the upper-left corner of this Jupyter notebook.)

generated_image = generate_noise_image(content_image)
imshow(generated_image[0])

model = load_vgg_model("pretrained-model/imagenet-vgg-verydeep-19.mat") # load VGG model

To get the program to compute the content cost, we will now assign a_C and a_G to be the appropriate hidden layer activations. We will use layer conv4_2 to compute the content cost. The code below does the following:

1. Assign the content image to be the input to the VGG model.
2. Set a_C to be the tensor giving the hidden layer activation for layer "conv4_2".
3. Set a_G to be the tensor giving the hidden layer activation for the same layer.
4. Compute the content cost using a_C and a_G.

# Assign the content image to be the input of the VGG model.  
sess.run(model['input'].assign(content_image))

# Select the output tensor of layer conv4_2
out = model['conv4_2']

# Set a_C to be the hidden layer activation from the layer we have selected
a_C = sess.run(out)

# Set a_G to be the hidden layer activation from same layer. Here, a_G references model['conv4_2'] 
# and isn't evaluated yet. Later in the code, we'll assign the image G as the model input, so that
# when we run the session, this will be the activations drawn from the appropriate layer, with G as input.
a_G = out

# Compute the content cost
J_content = compute_content_cost(a_C, a_G)

Note: At this point, a_G is a tensor and hasn't been evaluated. It will be evaluated and updated at each iteration when we run the Tensorflow graph in model_nn() below.

# Assign the input of the model to be the "style" image 
sess.run(model['input'].assign(style_image))

# Compute the style cost
J_style = compute_style_cost(model, STYLE_LAYERS)

Exercise: Now that you have J_content and J_style, compute the total cost J by calling total_cost(). Use alpha = 10 and beta = 40.

### START CODE HERE ### (1 line)
J = total_cost(J_content,J_style,alpha=10,beta=40)
### END CODE HERE ###

You'd previously learned how to set up the Adam optimizer in TensorFlow. Lets do that here, using a learning rate of 2.0. See reference

# define optimizer (1 line)
optimizer = tf.train.AdamOptimizer(2.0)

# define train_step (1 line)
train_step = optimizer.minimize(J)

Exercise: Implement the model_nn() function which initializes the variables of the tensorflow graph, assigns the input image (initial generated image) as the input of the VGG16 model and runs the train_step for a large number of steps.