YOLO-CAPTCHA

KO | EN

gitlite — search

#python #hacktoberfest #ai #llm #go #docker #macos #java #cli #kubernetes #security #mcp #openai #open-source

(★ 12)

本项目是一个基于 YOLO11n 架构的轻量化验证码识别模型。

YOLO-CAPTCHA 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

---

# YOLO11-Captcha-Recognition | 验证码识别微模型

[English](#english) | [中文](#chinese)

---

## 🇨🇳 中文说明

### 项目简介
本项目是一个基于 **YOLO11n** 架构的轻量化验证码识别模型。专门针对验证码场景中的 **文字 (Word)** 和 **图标 (Icon)** 进行优化。该模型具有极小的体积和极高的推理速度，非常适合集成到自动化工具中。

### 🚀 性能指标 (Metrics)
得益于 YOLO11 架构的强大性能，本模型在私有数据集上达到了以下表现：
- **mAP50**: `0.995` (近乎完美的识别率)
- **推理速度**: 约 `0.9ms` (在 RTX 5060 Laptop 级别显卡上)
- **模型大小**: 约 `5.5MB`

### 🖼️ 效果演示
![演示图片](https://raw.githubusercontent.com/ZKY-DW-Wait-me/YOLO-CAPTCHA/main/assets/result1.jpg)
*(上图展示了模型对中文字符的具体定位与分类能力，置信度均在 0.9 以上)*

### 📈 训练曲线
![训练结果](https://raw.githubusercontent.com/ZKY-DW-Wait-me/YOLO-CAPTCHA/main/assets/results.png)
*(Loss 曲线平稳收敛，各项指标快速爬升至 0.99 左右)*

### 🛠️ 快速上手
1. **安装环境**：
   ```bash
   pip install ultralytics torch
   ```
2. **推理代码**：
   ```python
   from ultralytics import YOLO

# 加载模型
   model = YOLO("weights/best.pt")

# 预测结果
   results = model.predict(source="your_image.jpg", save=True, conf=0.5)
   ```

---

## 🇺🇸 English Description

### Introduction
This project features a lightweight captcha recognition model based on the **YOLO11n** architecture. It is specifically optimized for detecting **Words** and **Icons** in captcha-style images. The model is extremely compact and fast, making it ideal for automation integration.

### 🚀 Key Performance
- **mAP50**: `0.995`
- **Inference Speed**: ~`0.9ms` (Tested on RTX 5060 Laptop GPU)
- **Model Size**: ~`5.5MB`

### 🖼️ Showcase
![Demo](https://raw.githubusercontent.com/ZKY-DW-Wait-me/YOLO-CAPTCHA/main/assets/result1.jpg)
*(High-confidence detection examples for Chinese characters and icons)*

### 📈 Training Logs
![Validation Results](https://raw.githubusercontent.com/ZKY-DW-Wait-me/YOLO-CAPTCHA/main/assets/ing robust convergence and high precision)*

### 🛠️ Quick Start
1. **Prerequisites**:
   ```bash
   pip install ultralytics torch
   ```
2. **Prediction**:
   ```python
   from ultralytics import YOLO

model = YOLO("weights/best.pt")
   results = model.predict(source="your_image.jpg", save=True, conf=0.5)
   ```

## ⚖️ License / 声明
- **LICENSE**: MIT
- **Disclaimer**: This project is for educational and research purposes only. Any illegal use is strictly prohibited.
- **免责声明**: 本项目仅供学习交流使用，严禁用于任何违法违规行为。

---

YOLO-CAPTCHA

YOLO-CAPTCHA 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

YOLO-CAPTCHA

YOLO-CAPTCHA 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README