Performance Improvement of Wild Flower Boundary Extraction in Complex Background

Abstract:

The performance of an automatic system for extracting flower boundaries for ten different types of wild flowers has been improved. The proposed flower boundary extraction method consists of three major procedures: the detection of four edge points, boundary tracing and performance improvement part. The flower boundary extraction part involves a new technique for automatically identifying the boundary of a flower in an image. An Intelligent Scissor algorithm is applied for boundary tracing. The color gradient magnitude and Canny edge detection are analyzed and included as the cost terms of the Intelligent Scissor algorithm. The color gradient magnitude cost term is implemented so that it can act directly on the three components of the color image. For the third procedure, we implement performance improvement. The main advantage of the proposed method was that when the program detected the wrong four edge points, using the mouse the correct positions could be clicked. The proposed method was applied to 500 photos of 10 different flowers, with 50 photos of each flower all in a complex background. The experimental results showed an extraction rate of 79.4%, which was better than before.

• Periodical: Applied Mechanics and Materials (Volumes 284-287)
• Pages: 2407-2410
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.284-287.2407
• Online since: January 2013
• Authors: Xiang Hua Li, Hyo Haeng Lee, Kwang Woo Chung, Kwang Seok Hong
• Keywords: Boundary Extraction, Canny Edge Detection, Four Edge Points, Maximum Gradient Magnitude

References

[1] Takeshi Saitoh and Toyohisa Kaneko. Automatic recognition of wild flowers. In Proc. of ICPR2000, volume 2, pages 507–510, Sep. (2000).

[2] Nagy, G.; Jie Zou; , Interactive visual pattern recognition, Pattern Recognition, 2002. Proceedings. 16th International Conference on , vol. 2, no., pp.478-481 vol. 2, (2002).

[3] Jie Zou; Nagy, G, Evaluation of model-based interactive ower recognition, Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on , vol. 2, no., pp.311-314 Vol. 2, 23-26 Aug. (2004).

DOI: 10.1109/icpr.2004.1334185

[4] Mei-Yan Jin, Kwang-Seok Hong, A Study on Improved Automatic Flower Boundary Detection, 3rd International Conference on Signal Acquisition and Processing(ICSAP 2011), (2011).

[5] Eric N. Mortensen and William A. Barrett. Intelligentscissors for image composition. In Proc. of SIGGRAPH, 1995, pages 191–198, Aug. (1995).