Using Convolution neural network for semantic image segmentation

Question

I am naive to python and this is my first attempt so please bear with me if my question seems to be too basic. I would like to use convolutional neural network to perform semantic segmentation to 9 satellite images. I have so far been successful in importing the images one by one and converting them to grayscale.

I would like to do the following process :

convolution- 16 filters, 33 filter size pooling- 22 filter size output- 4 classes testing and validation 80:20

any lead on this could be helpful. kindly guide !

Long Luu · Accepted Answer · 2020-08-05 05:10:40Z

2

Here is the code, only the model part:

import tensorflow as tf
from tensorflow.keras import Conv2D, Dense, Flatten, MaxPool2D

# Declare your desired things here
num_filter=32
kernel_size=(3,3)
strides=(1,1)
padding="valid"
input_shape=(width,height,channel)

model = tf.keras.Sequential([
   Conv2D(num_filter, kernel_size, strides=strides, input_shape=input_shape),
   MaxPool2D(),
   Flatten(),
   Dense(4, activation="softmax")
])

Here is a useful link: ConvNet TensorFlow guide

edited Aug 5, 2020 at 5:10

answered Aug 5, 2020 at 5:03

Long Luu

2,7793 gold badges21 silver badges53 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

user157522 Over a year ago

Thank you so much! Kindly guide. where can I enter my input images ?

Long Luu Over a year ago

You can use model.fit(x_train, y_train...), in the link there is also the tutorial.

Long Luu Over a year ago

If you find my answer is acceptable, consider stackoverflow.com/help/someone-answers

Collectives™ on Stack Overflow

Using Convolution neural network for semantic image segmentation

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related