# Intrusion Detection System based on YOLOv5 and DeepSort
**Update 2024.05.03**
---
## Project Introduction
Using PyQt5 to add a visual detection interface for YOLOv5, and implementing simple interface switching and area intrusion detection, as follows:
**Features:**
* UI interface and logical code are separated
* Supports user-selected models
* Outputs detection results and corresponding information
* Supports video, and camera detection
* Supports recording and querying intrusion detection records
* Supports video pause and continue detection
* Supports area intrusion, custom area, and counting.
## Quick Start
**Environment and Related File Configuration:**
- Set up the environment according to the `requirements`.
- Install PyQt5 and make sure to install and configure it.
- Download or train a model and put the `.pt` file into the `weights` folder. You can choose your own weight file, and the program will open the `weights` folder by default.
- The YOLOv5 release version used is v5.0: [Releases · ultralytics/yolov5 (github.com)](https://github.com/ultralytics/yolov5/releases)
- Need to configure the external tools Qtdesigner and PYUIC in PyCharm.
**How to Use:**
- Run `detect_logicwd.py` directly to enter the detection interface.
```shell script
python detect_logicwd.py
```
## Usage
1. After launching `detect_logicwd.py`, the software interface will be displayed as shown below:
![homepage](README.assets/homepage.png)
The system interface is divided into the left function selection, the right object detection object information, and the middle detection result screen.
2. First, select an available yolov5 weight file in the **功能选择** section, and then click the **初始化模型** button to load the target detection model. The selection of weight models supports self-trained models, as long as they are placed in the corresponding weights folder and the corresponding comboBox text is added through PyQT.
![loadModel](README.assets/loadModel.png)
3. Subsequently, select **视频检测** and perform object detection by uploading local video. Click the **结束检测** button to terminate the current target detection.
1. Video detection:
![videoDetection](README.assets/videoDetection.png)
4. In **视频检测**, you can draw a polygonal area. When an object enters the polygonal area, the category of the object will be detected (the judgment criterion for object entry is the relative position relationship between the center position of the detected box recognized by the target detection and the polygon).
1. Drawing operation: Check **区域入侵**, click **绘制区域**, click the left mouse button in the **检测结果** area to create the coordinates of the polygon. When the number of created points is greater than or equal to 2, the invasion detection algorithm will automatically start, and only the objects that enter the drawn area will be detected and related detection information will be output, while the objects not outside the area or recognized will not have any output.
![drawArea](README.assets/drawArea.png)
2. Upload the json format coordinate file: Prepare the file in advance, the content is organized as follows, (xn, yn) represents the coordinate information of a point of the polygon, and the coordinate system takes the upper left corner area of **检测结果** in the software interface as the origin.
![jsonFile](README.assets/jsonFile.png)
Check **区域入侵**, select the uploaded json file, and the invasion detection algorithm will be enabled.
![jsonDetection](README.assets/jsonDetection.png)
3. Query the records: You can use the date to query the intrusion records of the day
![queryRecords](README.assets/queryRecords.png)
4. In addition to local videos, real-time video streaming detection is also supported by providing the corresponding video streaming information. Both mouse drawing and uploading area methods are supported for invasion detection, which are not demonstrated due to limitations in computer performance.
## reference
- Github link1: https://github.com/jaycheney/YOLOv5-Intrusion-Detection-System
- Github link2: https://github.com/Sharpiless/yolov5-deepsort/
---
If you find this repository useful, please consider giving it a star ⭐️! Thanks!
没有合适的资源?快使用搜索试试~ 我知道了~
基于 Yolov5 和 Deepsort 的 GUI 入侵检测系统
共197个文件
pyc:79个
py:55个
yaml:26个
1.该资源内容由用户上传,如若侵权请联系客服进行举报
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
版权申诉
0 下载量 102 浏览量
2024-05-15
17:11:34
上传
评论
收藏 99.54MB ZIP 举报
温馨提示
使用 PyQt5 为 YOLOv5 添加可视化检测接口,实现简单的接口切换和区域入侵检测,具体如下: 特征: UI界面和逻辑代码是分开的 支持用户选择的型号 输出检测结果和相应信息 支持视频和摄像头检测 支持记录和查询入侵检测记录 支持视频暂停和继续检测 支持区域入侵、自定义区域和计数
资源推荐
资源详情
资源评论
收起资源包目录
基于 Yolov5 和 Deepsort 的 GUI 入侵检测系统 (197个子文件)
intrusionData.db 48KB
Dockerfile 821B
.gitattributes 66B
.gitignore 47B
.gitkeep 0B
Yolov5-Deepsort-main.iml 490B
train.jpg 59KB
ruqin.json 102B
test.json 66B
LICENSE 1KB
README.md 4KB
README.md 2KB
README.md 65B
1.mp4 43.83MB
queryRecords.png 747KB
videoDetection.png 577KB
drawArea.png 551KB
jsonDetection.png 519KB
loadModel.png 66KB
homepage.png 40KB
jsonFile.png 26KB
yolov5s.pt 14.11MB
datasets.py 44KB
detect_ui.py 31KB
general.py 28KB
detect_logicwd.py 28KB
plots.py 20KB
wandb_utils.py 16KB
common.py 16KB
yolo.py 13KB
torch_utils.py 12KB
json_logger.py 11KB
loss.py 9KB
metrics.py 9KB
kalman_filter.py 8KB
autoanchor.py 7KB
export.py 7KB
linear_assignment.py 6KB
train.py 6KB
nn_matching.py 5KB
line_draw.py 5KB
experimental.py 5KB
google_utils.py 5KB
track.py 5KB
io.py 4KB
tracker.py 4KB
activations.py 4KB
evaluation.py 3KB
deep_sort.py 3KB
original_model.py 3KB
model.py 3KB
iou_matching.py 3KB
Mouse_draw.py 3KB
test.py 2KB
preprocessing.py 2KB
feature_extractor.py 2KB
intrusionData.py 1KB
draw.py 1KB
resume.py 1KB
restapi.py 1KB
parser.py 1000B
detection.py 825B
log_dataset.py 800B
tools.py 734B
tracker.py 638B
__init__.py 500B
log.py 463B
asserts.py 316B
example_request.py 299B
evaluate.py 293B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
datasets.cpython-311.pyc 72KB
general.cpython-311.pyc 45KB
detect_ui.cpython-311.pyc 40KB
common.cpython-311.pyc 36KB
plots.cpython-311.pyc 36KB
datasets.cpython-37.pyc 34KB
yolo.cpython-311.pyc 23KB
general.cpython-37.pyc 23KB
torch_utils.cpython-311.pyc 22KB
common.cpython-36.pyc 19KB
common.cpython-37.pyc 19KB
plots.cpython-37.pyc 16KB
metrics.cpython-311.pyc 15KB
autoanchor.cpython-311.pyc 13KB
experimental.cpython-311.pyc 12KB
torch_utils.cpython-37.pyc 11KB
yolo.cpython-37.pyc 11KB
kalman_filter.cpython-311.pyc 9KB
nn_matching.cpython-311.pyc 8KB
tracker.cpython-311.pyc 8KB
metrics.cpython-37.pyc 8KB
linear_assignment.cpython-311.pyc 8KB
track.cpython-311.pyc 7KB
共 197 条
- 1
- 2
资源评论
hakesashou
- 粉丝: 4480
- 资源: 1260
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功