隨機(jī)寬線檢測(cè)方法
doi: 10.11999/JEIT170296
基金項(xiàng)目:
國家自然科學(xué)基金 (61401504)，博士后科學(xué)基金(2014M 562562)
Randomized Wide Line Detector
Funds:
The National Natural Science Foundation of China (61401504), China Postdoctoral Science Foundation (2014M 562562)
-
摘要: 為消除基本寬線檢測(cè)算子中的冗余計(jì)算量，提高算法的運(yùn)算速度，該文提出一種寬線算子的快速實(shí)現(xiàn)方法隨機(jī)移動(dòng)寬線算子。基本寬線算子采取逐像素移動(dòng)模板的方式檢測(cè)圖像中的寬線特征，與之不同，隨機(jī)移動(dòng)寬線算子在檢測(cè)時(shí)，隨機(jī)地在圖像中放置檢測(cè)模板，并根據(jù)當(dāng)前像素類型采用啟發(fā)式的準(zhǔn)則確定模板移動(dòng)的策略，從而加快了模板移動(dòng)速度，較好地消除了基本寬線檢測(cè)算法中的冗余運(yùn)算；在此基礎(chǔ)上，提出了兩種提前結(jié)束條件，可根據(jù)檢測(cè)情況提前結(jié)束循環(huán)，進(jìn)一步節(jié)省了運(yùn)算量。利用測(cè)試圖像對(duì)快速算子進(jìn)行了實(shí)驗(yàn)分析，結(jié)果表明，隨機(jī)移動(dòng)寬線算子在取得相當(dāng)檢測(cè)性能的同時(shí)，提高了基本寬線算子的運(yùn)算速度。
-
關(guān)鍵詞:
- 圖像處理 /
- 寬線特征 /
- 寬線算子 /
- 隨機(jī)移動(dòng) /
- 冗余計(jì)算
Abstract: To eliminate computation redundancy and improve speed of the basic wide line detector, a fast implementation, named randomized moving wide line detector, is proposed. Instead of moving the mask pixel by pixel to detect wide lines as did in the basic implementation, the randomized moving wide line detector places the mask in the image randomly, and then determines the mask moving strategy heuristically according to the current pixel. In this way, the mask moving is accelerated, leading to obvious decrease of computational redundancy in the basic detector. Furthermore, two early termination conditions are proposed to break out of the detecting loop based on the detection situation of wide lines. Testing images are adopted for performance evaluation of the randomized moving wide line detector. Experimental results demonstrate that the proposed detector accelerates the basic wide line detector significantly while keeping its detection performance unaffected.-
Key words:
- Image processing /
- Wide line feature /
- Wide line detector /
- Randomized moving /
- Computational redundancy
-
LINDEBERG T. Edge detection and ridge detection with automatic scale selection[J]. International Journal of Computer Vision, 1998, 30(2): 117-154. doi: 10.1023/A: 1008097225773. JACOB M and UNSER M. Design of steerable filters for feature detection using Canny-Like criteria[J]. IEEE Transaction on Pattern Analysis Machine Intelligence, 2004, 26(8): 1007-1019. doi: 10.1109/TPAMI.2004.44. EBERLY D, GARDNER R, MORSE B, et al. Ridges for image analysis[J]. Journal of Mathematical Imaging and Vision, 1994, 4(4): 353-373. doi: 10.1007/BF01262402. AGGARWAL N and KARL W C. Line detection in images through regularized Hough transform[J]. IEEE Transactions on Image Processing, 2006, 15(3): 582-591. doi: 10.1109/TIP. 2005.863021. XU Zezhong, SHIN Bok Suk, and KLETTE Reinhard. Closed form line-segment extraction using the Hough transform[J]. Pattern Recognition, 2015, 48(12): 4012-4023. doi: 10.1016/j. patcog.2015.06.008. LOPEZ-MOLINA C, VIDAL-DIEZ DE ULZURRUN G, BAETENS J M, et al. Unsupervised ridge detection using second order anisotropic Gaussian kernels[J]. Signal Processing, 2015, 116: 55-67. doi: 10.1016/j.sigpro.2015.03. 024. HU Yangyang, ZHANG Wenqiang, LU Hong, et al. Wide line detection with water flow[C]. 2016 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Shenzhen, 2016: 1353-1355. PANDEY D, YIN Xiaoxia, WANG Hua, et al. Accurate vessel segmentation using maximum entropy incorporating line detection and phase-preserving denoising[J]. Computer Vision and Image Understanding, 2017, 155: 162-172. doi: 10.1016/j.cviu.2016.12.005. SUDESHNA S K and SANTI P M. Retinal blood vessel extraction using tunable bandpass filter and fuzzy conditional entropy[J]. Computer Methods and Programs in Biomedicine, 2016, 133: 111-132. doi: 10.1016/j.cmpb.2016. 05.015. LI Shuxiao, CHANG Hongxing, and ZHU Chengfei. Fast curvilinear structure extraction and delineation using density estimation[J]. Computer Vision and Image Understanding, 2009, 113(6): 763-775. doi: 10.1016/j.cviu.2009.01.003. LIU L, ZHANG D, and YOU J. Detecting wide lines using isotropic nonlinear filtering [J]. IEEE Transactions on Image Processing, 2007, 16(6): 1584-1595. doi: 10.1109/TIP.2007. 894288. LIU L, NGADI M O, PRASHER S O, et al. Objective determination of pork marbling scores using the wide line detector[J]. Journal of Food Engineering, 2012, 110: 497-504. doi: 10.1016/j.jfoodeng.2011.11.008. CRISTIAN V and SERGIU N. Detecting curvilinear featuresusing structure tensors[J]. IEEE Transactions on Image Processing, 2015, 24(11): 3874-3887. doi: 10.1109/TIP. 2015.2447451. SMITH S M and BRADY J M. SUSANA new approach to low level image processing[J]. International Journal of Computer Vision, 1997, 23(1): 45-78. doi: 10.1023/A: 1007963824710. KNUTH D E, MORRIS J H, and PRATT V R. Fast pattern matching in strings[J]. SIAM Journal of Computing, 1977, 6: 323-350. doi: 10.1137/0206024. BOYER R S and MOORE J S. A fast string searching algorithm[J]. Communications of the ACM, 1977, 20(10): 762-772. doi: 10.1145/359842.359859. ABDOU I E and PRATT W K. Quantitative design and evaluation of enhancement/thresholding edge detectors[J]. Proceedings of the IEEE, 1979, 67(5): 753-763. doi: 0.1109/ PROC.1979.11325. IGOR S and ALEKSEJ A. Convolutional neural network based automatic object detection on aerial images[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 14(5): 740-744. doi: 10.1109/LGRS.2016.2542358. RODRIGO F N, ROBERTO A L, and RUBENS C M. Fingerprint liveness detection using convolutional neural networks[J]. IEEE Transactions on Information Forensics and Security, 2016, 11(6): 1206-1213. doi: 10.1109/TIFS. 2016.2520880. -
計(jì)量
- 文章訪問數(shù): 963
- HTML全文瀏覽量: 151
- PDF下載量: 155
- 被引次數(shù): 0
下載:
