研究目的
To build an image processing flame detection system using image enhancement, segmentation, and filtering methods applied in LabVIEW for early fire detection.
研究成果
The flame detection system using HSV color filtering in LabVIEW achieves up to 98% accuracy in detecting flames under controlled conditions. However, it struggles with false positives from similar light sources. Future work should focus on improving discrimination between flames and other luminous objects.
研究不足
The system cannot accurately distinguish between real flames and other light sources (e.g., LED lights, bulbs) due to reliance on luminance properties. Performance is optimal in low-light conditions but less effective in high-light environments.
1：Experimental Design and Method Selection:
The system uses image processing techniques including image enhancement (brightness and contrast adjustment), segmentation in HSV color space, grayscale conversion, binary image processing, and blob detection. LabVIEW software with Vision Assistant module is employed for implementation.
2：Sample Selection and Data Sources:
Real-time images captured by a Logitech C170 webcam under various lighting conditions (morning, afternoon, evening, night, with lamp on/off) and distances from the flame.
3：List of Experimental Equipment and Materials:
Logitech C170 webcam, PC/laptop with 2.40 GHz processor and 3.00 GB RAM, LabVIEW 2014 software, Autodesk Inventor 2016 for mechanical design, servo motors for camera movement.
4：40 GHz processor and 00 GB RAM, LabVIEW 2014 software, Autodesk Inventor 2016 for mechanical design, servo motors for camera movement.
Experimental Procedures and Operational Workflow:
4. Experimental Procedures and Operational Workflow: Camera captures images at 30 fps with 1024x768 resolution. Images undergo pre-processing: brightness and contrast enhancement, HSV color filtering, grayscale conversion, binary processing, and blob detection. If blob size exceeds 2000 pixels, a region of interest (ROI) is marked, and an alarm is triggered.
5：Data Analysis Methods:
Analysis of detected pixel counts under different conditions (distance, lighting, object type) to evaluate system accuracy and performance.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
