[go: up one dir, main page]
Scribd
Upload
3 views8 pages

SDLC DeepLearning DigitRecognition

The document outlines the Software Development Life Cycle (SDLC) for a Handwritten Digit Recognition App using deep learning techniques. It covers phases including requirement analysis, design, development, testing, and deployment, detailing the use of a CNN model with the MNIST dataset. The project aims for high accuracy and quick response times, with plans for future enhancements and maintenance strategies.

Uploaded by

Rohan K
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
3 views8 pages

SDLC DeepLearning DigitRecognition

The document outlines the Software Development Life Cycle (SDLC) for a Handwritten Digit Recognition App using deep learning techniques. It covers phases including requirement analysis, design, development, testing, and deployment, detailing the use of a CNN model with the MNIST dataset. The project aims for high accuracy and quick response times, with plans for future enhancements and maintenance strategies.

Uploaded by

Rohan K
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 8

SDLC for Deep Learning Software

— Digit Recognition App


A Real-World AI Project Lifecycle
Presented by: [Your Name]
Date: [Insert Date]
Introduction to SDLC
• SDLC (Software Development Life Cycle):
Structured approach to software creation.

• Phases Covered:
• 1. Requirement Analysis
• 2. Design
• 3. Development
• 4. Testing
• 5. Deployment
Project Overview
• Project: Handwritten Digit Recognition

• Goal: Recognize digits (0–9) from images using


a CNN.
• Dataset: MNIST (70,000 grayscale images)

• Tech Stack:
• - Python
• - TensorFlow/Keras
Requirement Analysis
• Functional Requirements:
• - Upload image
• - Predict digit
• - Display result

• Non-functional Requirements:
• - Accuracy > 95%
• - Response time < 2 seconds
Design Phase
• Model Architecture: CNN with 2 Conv layers,
ReLU, MaxPooling, Flatten, Dense layers

• System Design:
• Frontend → Flask API → Deep Learning Model

• Flow: User → Image Upload → API → CNN


Model → Prediction Output
Development Phase
• Model Code (Keras):
• model = Sequential([
• Conv2D(32, (3,3), activation='relu',
input_shape=(28,28,1)),
• MaxPooling2D(2,2),
• Flatten(),
• Dense(128, activation='relu'),
• Dense(10, activation='softmax')
• ])
Testing Phase
• Manual Test Cases:
• | Input | Expected Output | Status |
• |-------|-----------------|--------|
• | Image of '3' | Output: 3 | ✅ |
• | Letter 'A' | Output: Error | ✅ |

• Automated Test:
• assert predict_digit(test_image) == 7
Deployment & Maintenance
• Deployment:
• - Flask app hosted on Heroku/Render
• - Model saved as .h5 file

• Maintenance:
• - Retrain with new data
• - Optimize for mobile

• Future Enhancements:

You might also like