Automated-E-commerce-Compliance-Verification-System

Automated E-commerce Compliance Verification System [inorder to clone this project please check my flipkart clone in my github repo we use my flipkart clone backend api endpoint for crawling data.I even given my deployed flipkart clone in render but the issue is a project get shutdown after 15min of inactiveness in render so it get restarted when we freshly open the project so it takes time so if you need better experience clone my flipkart cline use local backend endpoint] 🚀 Project Overview An intelligent compliance validation framework that leverages computer vision and distributed computing to automatically verify product compliance across e-commerce platforms. This system performs real-time regulatory compliance checking using advanced optical character recognition (OCR) and multi-layered validation protocols.

🔬 System Architecture Distributed Processing Pipeline The system implements a horizontally scalable architecture with dual-node processing:

System A: Data Acquisition & Orchestration Node

System B: Computer Vision & Compliance Analysis Node { Before proceeding into execution you have to do one thing, keep both systems on static ip’s so communication wont disturb as ip does change in time being. Replace system B ip address in systemA code. And replace frontend port in both system codes as i explicitly allowd only frontend port through cors } Technical Infrastructure Multi-threaded Data Processing: Parallel execution for optimized throughput

Intelligent Caching Mechanism: Redis-inspired memory caching for performance optimization

RESTful Microservices: Decoupled architecture for independent scaling

Real-time Image Processing: Advanced OCR with contextual analysis

🎯 Core Features

1. Automated Compliance Validation Dual-Layer Verification Protocol:

Layer 1: Universal compliance flags (Manufacturer details, MRP, Quantity, etc.)

Layer 2: Category-specific regulatory requirements

1. Intelligent OCR Processing EasyOCR Integration for high-accuracy text extraction

Multi-modal Data Analysis: Combines image text + product metadata

Contextual Keyword Recognition with semantic understanding

1. Distributed Computing Parallel Processing Pipeline for scalable operations

Network-Optimized Communication between systems

Load-Balanced Task Distribution

1. IoT Integration Ready Raspberry Pi Camera Endpoint for physical product verification

Real-time Image Analysis for instant compliance checking

🛠 Technical Implementation Compliance Validation Matrix Universal Compliance Layer (Layer 1)

{ “Manufacturer/Importer Name & Address”: False, “MRP”: False, “Net Quantity”: False, “Date of Manufacture/Expiry”: False, “Country of Origin”: False, “Consumer Care Details”: False } Category-Specific Compliance (Layer 2) Electronics: BIS Certification, Warranty, Power Specifications

Grocery: FSSAI, Expiry Dates, Nutritional Information

Mobiles: SAR Value, Battery Capacity, Charger Specifications

Beauty: Ingredients, Manufacturing Licenses, Usage Instructions

And 6+ additional categories with specialized compliance requirements

🌐 API Endpoints System A (Data Acquisition & Orchestration) GET /crawl-and-send - Initiates product crawling and distributed processing

GET /get-results - Retrieves comprehensive compliance results

GET /get-result/ - Product-specific compliance report

GET /stats - System performance and compliance statistics

POST /clear-data - Cache management and data reset

System B (Computer Vision & Analysis) POST /process-product - Main compliance validation endpoint

POST /analyze-image - Raspberry Pi camera image analysis endpoint

🔄 Workflow Pipeline Data Ingestion: Automated product crawling from e-commerce API

Distributed Processing: Parallel data transfer to analysis node

Multi-modal Analysis:

Image preprocessing and OCR text extraction

Metadata pattern recognition

Compliance flag validation

Intelligent Caching: Results stored in-memory for optimized performance

Real-time Reporting: JSON-based compliance reports via REST API

🚀 Performance Optimizations Distributed Load Handling: Splits computational workload across multiple systems

Memory Caching Layer: Prevents redundant processing of previously analyzed products

Network-Efficient Communication: Optimized data transfer between nodes

Scalable Architecture: Designed for horizontal scaling to 8+ processing threads

🔮 Future Enhancements Machine Learning Integration for improved pattern recognition

Blockchain Verification for tamper-proof compliance records

Cloud-Native Deployment with container orchestration

Real-time Dashboard with compliance analytics

Multi-platform E-commerce Support beyond Flipkart clone

💡 Innovation Highlights First-of-its-kind automated compliance framework for e-commerce

Hybrid validation approach combining computer vision and metadata analysis

Distributed architecture solving real-time processing challenges

IoT-ready implementation for physical product verification

Category-adaptive compliance rules with dynamic validation matrix

🏆 Business Impact Prevents fraudulent product listings through automated verification

Ensures regulatory compliance across multiple product categories

Reduces manual inspection overhead by 90%+

Scalable solution capable of handling millions of product listings

Real-time compliance monitoring for dynamic e-commerce environments

FRONTEND: I didnt focus much on frontend but kept it pretty descent.You can modify it based on your interest as i have given all the endpoints. How to use frontend: 1.run my html code(after running of my python codes in 2 systems) 2.press start compilance button(wait until loader loads, in background you can check logs of my both python scripts wether crawling and ocr is performing or not) 3.after loader stops rotating automatically you see results on screen and explore them. 4.even after loader stops rotating if you dont see results press get results. 5.Note never press any button repeatedly becuase it increases load on system OUTPUT images: