Image segmentation basics in TensorFlow - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output shape of a simple segmentation model

Consider this TensorFlow model for image segmentation. What is the shape of the output tensor after the final layer?

import tensorflow as tf
model = tf.keras.Sequential([
  tf.keras.layers.InputLayer(input_shape=(128, 128, 3)),
  tf.keras.layers.Conv2D(16, 3, padding='same', activation='relu'),
  tf.keras.layers.Conv2D(1, 1, padding='same', activation='sigmoid')
])
output = model.output

TensorFlow

import tensorflow as tf
model = tf.keras.Sequential([
  tf.keras.layers.InputLayer(input_shape=(128, 128, 3)),
  tf.keras.layers.Conv2D(16, 3, padding='same', activation='relu'),
  tf.keras.layers.Conv2D(1, 1, padding='same', activation='sigmoid')
])
output = model.output
output_shape = output.shape

A (None, 126, 126, 16)

B (None, 126, 126, 1)

C (None, 128, 128, 16)

D (None, 128, 128, 1)

💡 Hint

Remember that padding='same' keeps the spatial dimensions the same.

The model input is 128x128 with 3 channels. The first Conv2D layer with padding='same' keeps the height and width at 128, outputting 16 channels. The second Conv2D layer also uses padding='same' and outputs 1 channel. Batch size is None (unspecified). So output shape is (None, 128, 128, 1).

🧠 Conceptual

intermediate

Purpose of activation function in segmentation output layer

In image segmentation models, why is the sigmoid activation function commonly used in the output layer for binary segmentation?

A It outputs values between 0 and 1 representing the probability of each pixel belonging to the target class.

B It normalizes the output to have zero mean and unit variance.

C It reduces the output size to half the input size.

D It increases the number of output channels to match the input channels.

💡 Hint

Think about what the model predicts for each pixel in binary segmentation.

Sigmoid activation outputs values between 0 and 1, which can be interpreted as probabilities for each pixel belonging to the target class or background. This is essential for binary segmentation tasks.

❓ Hyperparameter

advanced

Choosing loss function for multi-class segmentation

You are training a multi-class image segmentation model with 4 classes. Which loss function is most appropriate to use?

A Hinge loss

B Binary crossentropy

C Categorical crossentropy

D Mean squared error

💡 Hint

Consider the number of classes and the output format of the model.

Categorical crossentropy is designed for multi-class classification problems where each pixel belongs to exactly one of multiple classes. Binary crossentropy is for two classes only.

❓ Metrics

advanced

Evaluating segmentation quality with IoU

What does the Intersection over Union (IoU) metric measure in image segmentation?

A The ratio of correctly predicted pixels to total pixels in the image.

B The overlap between the predicted segmentation mask and the ground truth mask divided by their union.

C The average pixel intensity difference between prediction and ground truth.

D The time taken to segment an image.

💡 Hint

Think about how IoU compares two sets of pixels.

IoU measures how much the predicted mask and the true mask overlap relative to their combined area. It is a common metric to evaluate segmentation accuracy.

🔧 Debug

expert

Identifying cause of poor segmentation training

You trained a segmentation model but the training loss does not decrease and predictions are random. Which of these is the most likely cause?

A Using a sigmoid activation in the output layer but applying categorical crossentropy loss.

B Using a softmax activation in the output layer with categorical crossentropy loss.

C Normalizing input images to [0,1] range before training.

D Using data augmentation during training.

💡 Hint

Check if the activation and loss function match the problem type.

Sigmoid activation outputs independent probabilities per pixel channel, but categorical crossentropy expects a probability distribution (softmax output). This mismatch causes training failure.

Practice

1. What is the main goal of image segmentation in TensorFlow?

easy

A. To compress images without losing quality

B. To label each pixel in an image with a class

C. To generate random noise images

D. To classify the entire image into one category

Solution

1. Step 1: Understand image segmentation purpose

   Image segmentation assigns a class label to every pixel, dividing the image into meaningful parts.

2. Step 2: Compare with other tasks

   Unlike classification, which labels the whole image, segmentation works at pixel level.

3. Final Answer:

   To label each pixel in an image with a class -> Option B

4. Quick Check:

   Image segmentation = pixel-wise labeling [OK]

Hint: Segmentation means labeling pixels, not whole images [OK]

Common Mistakes:

• Confusing segmentation with image classification
• Thinking segmentation compresses images
• Believing segmentation creates random images

2. Which TensorFlow layer is commonly used to extract features for image segmentation?

easy

A. Convolutional layer

B. Dense layer

C. Dropout layer

D. Flatten layer

Solution

1. Step 1: Identify feature extraction layers

   Convolutional layers scan images to find patterns and features useful for segmentation.

2. Step 2: Understand other layers' roles

   Dense layers are for classification, dropout prevents overfitting, flatten reshapes data but doesn't extract features.

3. Final Answer:

   Convolutional layer -> Option A

4. Quick Check:

   Feature extraction = convolutional layers [OK]

Hint: Convolutional layers find image features [OK]

Common Mistakes:

• Choosing dense layers for feature extraction
• Confusing flatten with convolution
• Ignoring convolution's role in images

3. What will be the shape of the output tensor after applying a softmax activation on a segmentation model predicting 3 classes for a 128x128 image?

medium

A. (128, 128, 3)

B. (3, 128, 128)

C. (128, 3, 128)

D. (128, 128)

Solution

1. Step 1: Understand output shape for segmentation

   Each pixel has a probability for each class, so output shape is height x width x classes.

2. Step 2: Apply to given image and classes

   Image size is 128x128, classes are 3, so output shape is (128, 128, 3).

3. Final Answer:

   (128, 128, 3) -> Option A

4. Quick Check:

   Output shape = (height, width, classes) [OK]

Hint: Output shape = image size + number of classes [OK]

Common Mistakes:

• Mixing class dimension position
• Forgetting classes dimension
• Using 2D shape instead of 3D

4. Given this TensorFlow code snippet for segmentation, what is the error?

model = tf.keras.Sequential([
  tf.keras.layers.Conv2D(16, 3, activation='relu', input_shape=(64, 64, 3)),
  tf.keras.layers.Dense(3, activation='softmax')
])

medium

A. Conv2D kernel size must be 1

B. Input shape is incorrect for images

C. Activation 'relu' is invalid

D. Dense layer cannot follow Conv2D without flattening

Solution

1. Step 1: Check layer sequence

   Conv2D outputs 4D tensor, Dense expects 2D input, so flattening is needed before Dense.

2. Step 2: Validate other options

   Input shape is valid for images, 'relu' is a valid activation, kernel size 3 is allowed.

3. Final Answer:

   Dense layer cannot follow Conv2D without flattening -> Option D

4. Quick Check:

   Dense after Conv2D needs flattening [OK]

Hint: Dense layers need flat input, add Flatten after Conv2D [OK]

Common Mistakes:

• Ignoring tensor shape mismatch
• Thinking kernel size must be 1
• Misunderstanding activation functions

5. You want to train a simple segmentation model in TensorFlow. Which data do you need to provide during training?

hard

A. Only images without labels

B. Images and image-level class labels

C. Images and their pixel-level masks

D. Random noise images

Solution

1. Step 1: Identify training data for segmentation

   Segmentation requires images and masks labeling each pixel's class to learn pixel-wise predictions.

2. Step 2: Compare with other data types

   Image-level labels are for classification, no labels means no supervision, noise images are irrelevant.

3. Final Answer:

   Images and their pixel-level masks -> Option C

4. Quick Check:

   Training segmentation = images + pixel masks [OK]

Hint: Segmentation training needs pixel masks, not just labels [OK]

Common Mistakes:

• Using only image-level labels
• Training without masks
• Confusing segmentation with classification data