HVC-Tunnel-Multimodal/Infrared-ai-research-lab
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 2 Wiki Activity
19 Commits 1 Branch 2 Tags
8e3206eb4f6c828b76cc889afcf5c1d297f07ebb
Commit Graph

19 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
SkyLi 8e3206eb4f docs: remove obsolete dataset download scripts from README 2026-03-29 00:29:47 +08:00
SkyLi 70892a17f9 chore: ignore src_docs and CLAUDE.md in docs directory 2026-03-29 00:24:30 +08:00
SkyLi be3b08ccd9 chore: remove unauthorized AI-generated license from README 2026-03-27 14:24:29 +08:00
SkyLi 17ee9f94da chore: remove obsolete project plan files 2026-03-27 12:32:11 +08:00
SkyLi 73797b98aa chore: remove obsolete NEW_PROJECT_PROMPT.md and update README 
- 删除NEW_PROJECT_PROMPT.md（项目已初始化完成，不再需要）
- 从README项目结构中移除已删除的convert_to_yolo.py
- 将convert_to_yolo.py命令替换为analyze_temp_features.py
- 更新FAQ中关于自定义数据集的说明
 2026-03-27 12:23:39 +08:00
SkyLi b4308d6cdc fix: update analyze_temp_features.py for new API 
- 移除对InfraredDetector的依赖，改用TempFeatureExtractor.detect_and_extract()
- 添加--device参数支持GPU/CPU切换
- 简化特征提取逻辑，直接使用detect_and_extract()返回的特征
- 添加空结果警告提示
 2026-03-27 12:21:22 +08:00
SkyLi 60b5b89a7c chore: remove legacy convert_to_yolo.py script 
新项目数据集已是完整标注的YOLO格式，无需旧的数据转换工具
 2026-03-27 12:14:54 +08:00
SkyLi 3bba21d83b refactor: merge _check_fault and _classify_fault into _analyze_fault 
一次调用返回(fault_status, fault_type)元组，简化诊断逻辑
 2026-03-27 12:10:27 +08:00
SkyLi 085cbc9950 fix: optimize diagnosis output format and fix inconsistency between fault_status and fault_type 
1. 输出格式优化:
   - faults中移除冗余的bbox和confidence字段
   - summary中fault_count改名为detection_count

2. fault_status与fault_type逻辑统一:
   - normal状态时fault_type为'none'，不再使用'suspect'
   - abnormal状态时才进行故障类型分类(hotspot/local_overheating/overheat)

3. 修复usage_example.py图像通道问题:
   - YOLO检测使用彩色图，温度提取时再转灰度图
 2026-03-27 12:03:48 +08:00
SkyLi febd9eea8d feat: add device parameter support for GPU/CPU switching in inference API 
- InfraredDetector.detect() 和 detect_and_draw() 新增 device 参数
- TempFeatureExtractor.detect_and_extract() 新增 device 参数
- FaultDiagnoser.diagnose() 和 batch_diagnose() 新增 device 参数
- batch_extract_features() 新增 device 参数
- 命令行 --device 选项支持指定推理设备
- detect_and_draw() 返回值改为 (image, n_detections) 避免重复检测
 2026-03-27 09:10:57 +08:00
SkyLi 9a5f6fe473 fix: fix multiple image duplicate loading issues 
- detect.py: detect_and_draw()中传给detect()的是已加载的img数组而非字符串路径
- usage_example.py: example3中复用gray_img数组而非重新加载
- analyze_temp_features.py: 复用已加载的gray_img数组进行检测
 2026-03-27 08:53:29 +08:00
SkyLi 10fb804c14 fix: fix image duplicate loading in diagnose function 
将cv2.imread从循环内移至循环前，避免对同一张图片重复加载
 2026-03-27 08:51:59 +08:00
SkyLi 70d624ac1d fix: correct project structure and file path errors in README 
- 修正项目根目录名称: 1BaseOnYOLO → 2BaseOnYOLO_InfraredData
- 移除不存在的脚本: download_dataset.py, retry_download.py
- 添加存在的脚本: analyze_temp_features.py
- 统一模型权重文件名: infrared_yolov8n.pt → best.pt
 2026-03-27 08:08:59 +08:00
SkyLi ff3dbbfd48 refactor: extract repeated sample image path to module constant 
Replace 6 inline string repetitions of the sample image path with
a single SAMPLE_IMAGE module-level constant for easier maintenance.
 2026-03-27 01:05:50 +08:00
SkyLi 787d17a870 chore: update project docs and scripts for new dataset 
- 更新README.md数据集描述（12类，6735训练图+765验证图）
- 更新模型性能指标（新12类模型结果）
- 删除已过时的download_dataset.py和retry_download.py脚本
- 更新usage_example.py使用正确的路径和模型名
 2026-03-27 00:56:10 +08:00
SkyLi 11c5e062c2 feat: add ROI temperature feature histogram analysis script 
Analyzes validation set images using YOLO detection, extracts ROI
temperature features, and plots histograms.
 2026-03-27 00:47:33 +08:00
SkyLi fc4a2eb674 fix: remove inline comments from .gitignore patterns 
Inline comments are not supported in gitignore - # only acts as a
comment when it appears at the beginning of a line. This caused
output/ to not be ignored because the pattern was matched literally.
 2026-03-27 00:46:55 +08:00
SkyLi 45fd5c5663 chore: update dataset config for 12-class training 
- 新数据集：12类电力设备红外图像（bushing, arrester, breaker等）
- 训练集：6735张，验证集：765张（删除2张越界标签图像）
- YOLOv8n模型训练50 epochs
- 验证结果：mAP@0.5=76.42%, mAP@0.5:0.95=61.64%
- 更新PROGRESS.md记录
 2026-03-27 00:16:21 +08:00
SkyLi fd1a5f4050 chore: initial commit for Infrared Equipment Detection System 
- Add YOLO-based infrared thermal image detection system
- Add dataset configuration (dataset.yaml) with 12 equipment classes
- Add training, validation and detection scripts
- Add temperature feature extraction and fault diagnosis modules
- Update README and CLAUDE.md with correct dataset paths
- Optimize .gitignore to track dataset.yaml while ignoring large data files
 2026-03-26 21:36:08 +08:00