Scribd
Upload
8 views5 pages

Project Progress Review Report (1) (Autorecovered)

Uploaded by

Aviral Srivastava
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
8 views5 pages

Project Progress Review Report (1) (Autorecovered)

Uploaded by

Aviral Srivastava
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

PROJECT PROGRESS REVIEW REPORT-I

PROJECT TITLE: Brain Tumor Detection using Machine Learning.

AREA OF WORK: Image Processing

SOFTWARE USED: Software (PYTHON 3.11.4)

PROJECT AIM & OBJECTIVES

Aim: This project aims to detect BRAIN TUMOR using MRI image through Python.

Project Objectives
Objective#1: P r e p r o c e s s i n g o f i m a g e s .
Objective#2: Model Training and Testing.
Objective#3: Creating Web App and Android App.

PROPOSED DESIGN METHODOLOGY: The proposed design methodology of brain tumor


detection using MRI Image through Python block diagram is shown below-
Block Diagram of Proposed Design Methodology for Brain Tumor Detection from MRI
Image Using Python and machine learning by Digital Image Processing.

PROGRESS IN WORK: During this course of time Objective# 1, Objective# 2, are completed
and Objective# 3 are under processing detailed in below:
Processing brain tumor images using a convolutional neural network (CNN) implemented in
Keras. Here is a short description. .of the code:

1. Import libraries: OpenCV (cv2), PIL (Image), NumPy, scikit-learn, Keras, etc. Import
libraries required for graphics, machine learning, and visualization, including matplotlib.

2. Data magnification: ImageDataGenerator is defined by magnification parameters such as


rotation, width/height setting, and horizontal rotation. This will apply the transformation to the
original image, creating additional training images, which can help increase dataset size and
improve model performance.

3. Morphological processing parameters: Defines the kernel size and iterations for the
morphological processing (erosion) applied to the image. This function can help improve features
in the image.

4. Loading and preprocessing the dataset: The code reads images from two grids, "no" (images
without tumors) and "yes" (images with tumors). It resizes the image to a reasonable size, uses
morphological operations, and stores the image and its associated text in separate objects.

5. Split the dataset: Use the train_test_split function in scikit-Learn to split the dataset into
training and test sets. The sample size is set to 20% of the entire dataset.

6. Data normalization and one-bit encoding: Normalize the input image using Keras'
normalization function. Use the to_categorical function to one-time encode text and convert them
to binary vectors.

7. Model Architecture: Use Keras to create sequential models. It has three layers of
convolutional layer followed by a max pooling layer. Next, a smoothing layer is added to convert
the 2D feature maps to 1D vectors. Then add a thick layer with corrected linear unit (ReLU)
activation and continuous output. Finally, the output method with softmax is added to divide the
image into two classes (tumor or no tumor).

8. Model compilation and training: Write a model using the power-down algorithm, Adam
optimizer, and truth-based index. Use the appropriate training model of the training process and
the training history is stored in the "history" variable.

9. Visualization of learning progress: Create two graphs to show training and actual use and
time wasted.

10. Prediction and evaluation: The model is used to predict the scores of the test. The predicted
results are converted into lecture notes. Build the confusion matrix using scikit-learn's
fusion_matrix function. Then use Matplotlib to plot and display the confusion matrix.

11.Use augmented data for further training: Use augmented data generated by
ImageDataGenerator to further train the model. This can help expand the model better by
exposing it to different training examples.

12. Save Model: Save the learned model to a file for future use.

PROGRESS REPORTED: 66%

Progress in work

Work Completed Work remaining

NAME AND SIGNATURE OF TEAM MEMBERS:


Member #1: Aviral Srivastava(195213)
Member #2: Shashank Kumar (2085260)
Member #3: Pramod Bind Modi(2085255)

(APPROVED/ NOT- APPROVED)

NAME AND SIGNATURE OF PROJECT SUPERVISOR

EVALUATION REMARKS BY PROJECT COORDINATOR(S):

You might also like