error-level-analysis

Image Tampering Detection using ELA and CNN

ELA 全称：Error Level Analysis ，汉译为“错误级别分析”或者叫“误差分析”。通过检测特定压缩比率重新绘制图像后造成的误差分布，可用于识别JPEG图像的压缩。

aim of this project is to give insight into authenticity of an image using ELA and metadata analysis based weather validation

Classifies a given image as authentic or tampered by doing two levels of analysis. Implemented using PyTorch.

Detects the authenticity of an image using Error Level Analysis and Convolutional Neural Networks.

Classifies a given aadhaar image to real or fake by doing two levels of analysis.

Web app for image forgery detection. Runs 12 published forensic techniques: ELA, Splicebuster noise analysis, double-JPEG and JPEG ghosts, copy-move, PRNU camera fingerprint, resampling, histogram, LSB steganalysis, metadata/C2PA and more. Reports what each test measured and its limits, never a black-box real-or-fake verdict. Includes demo images.

Employing Error Level Analysis (ELA) and Edge Detection techniques, this project aims to identify potential image forgery by analyzing discrepancies in error levels and abrupt intensity changes within images.

Developed an intelligent solution using OCR, QR code detection, and computer vision to extract and validate Aadhaar details from images. Applied preprocessing for rotated/skewed inputs, ensured fraud detection via pattern checks, and improved accuracy for secure, automated identity verification.

Edited Images Analyser

Image Forgery Detection using ELA and Deep Learning

This tool compares the original image to a recompressed version. This can make manipulated regions stand out in various ways. For example they can be darker or brighter than similar regions which have not been manipulated.

Detection of Human Edited Images using CNN, VGG16, Xception, ELA, Ensemble Learning.

Advanced deepfake detection and digital forensics platform using Error Level Analysis (ELA) and machine learning to verify digital media authenticity, detect image manipulation, analyze visual artifacts, identify tampering patterns, and provide users with clear, actionable insights into the authenticity and integrity of uploaded images with clarity

Image Forgery Detection using ELA and Deep Learning

A multi-layered AI forensic system combining ELA, CLIP, and Qwen2-VL to detect digital forgeries, deepfakes, and AI-generated synthetic media.

AI-powered image forensics tool that detects deepfakes and image tampering, then localizes forged regions using ELA + ResNet50.

Simple Tampered Image Detection using Error Level Analysis and Convolutional Neural Network with Flask

🎩 A comprehensive document authenticity verification tool with advanced image forensics and LLM-based text analysis.