pochih/CBIR：🏞 基于内容的图像检索（CBIR）系统

项目介绍

pochih/CBIR 是一个托管在 GitHub 上的开源项目，主要使用 Python 编写，目前获得 652 个 Star、170 次 Fork。

仓库全名: pochih/CBIR

README 摘要

Intro This repository contains a CBIR (content-based image retrieval) system Extract query image's feature, and retrieve similar ones from image database Image src Part1: Feature Extraction In this system, I implement several popular image features: color-based RGB histogram texture-based gabor filter shape-based daisy edge histogram HOG (histogram of gradient) deep methods VGG net Residual net all features are…

完整内容见 GitHub 原始 README 或本页。

常见问题

CBIR 用什么语言开发？

主要使用 Python 编写，目前有 652 个 Star。

如何获取 CBIR 的源码？

可以直接克隆仓库：git clone https://github.com/pochih/CBIR.git，默认分支为 master；安装与运行步骤见仓库 README。

CBIR 最近还在维护吗？

仓库最近一次更新为 3 天前，当前有 11 个未关闭 Issue；Star 数为 652，可作为社区活跃度的参考。

关于

GitHub 原始描述（英文）

🏞 A content-based image retrieval (CBIR) system

computer-vision daisy edges feature-fusion gabor histogram hog image-retrieval random-projection resnet vggnet

652 Stars
170 Forks
15 Watching

描述取自 GitHub 官方 API，保持仓库作者书写的原文。

简要分析

热度分

1007

价值分

443

活跃状态

活跃

GitHub

语言占比

Python

贡献者

pochih

Intro

This repository contains a CBIR (content-based image retrieval) system

Extract query image’s feature, and retrieve similar ones from image database

Image src

Part1: Feature Extraction

In this system, I implement several popular image features:

color-based
- RGB histogram
texture-based
- gabor filter
shape-based
deep methods
- VGG net
- Residual net

all features are modulized

Feature Fusion

Some features are not robust enough, and turn to feature fusion

fusion.py

Dimension Reduction

The curse of dimensionality told that vectors in high dimension will sometimes lose distance property

Random Projection

Part2: Evaluation

CBIR system retrieves images based on feature similarity

Robustness of system is evaluated by MMAP (mean MAP), the evaluation formulas is refer to here

image AP : average of precision at each hit
- depth=K means the system will return top-K images
- a correct image in top-K is called a hit
- AP = (hit1.precision + hit2.precision + … + hitH.precision) / H
class1 MAP = (class1.img1.AP + class1.img2.AP + … + class1.imgM.AP) / M
MMAP = (class1.MAP + class2.MAP + … + classN.MAP) / N

Implementation of evaluation can found at evaluate.py

My database contains 25 classes, each class with 20 images, 500 images in total, depth=K will return top-K images from database