基于几何证据收集的物体检测

doi:10.11896/j.issn.1002-137X.2016.10.057

摘要/Abstract

摘要： 人造物体通常具有非常稳定的形状特征,这种几何上较为持久和稳定的属性为物体识别提供了证据,并且它们相对于外观特征、颜色特征或灰度梯度特征等更具有稳定性和分辨力。基于形状特征对物体进行识别的难点在于物体的颜色、光照、尺寸、位置、姿态和背景干扰总是在不断变化的,且不可能事先预测到所有可能的环境。这种物体本身和环境的多样性使得基于几何形状的物体识别成为一个非常具有挑战性的难题。通过定义一种基于形状模板、对边缘线段化之后的图像进行几何证据筛选、收集和组合判断的方法,实现从背景环境中精确找到目标物体,并能够指出组成物体的线段的语义属性。该方法的实质是解一个全局最优的组合优化问题。虽然全局最优组合优化问题看似复杂度很高,但它无需定义复杂的特征向量,无需高代价的样本训练过程,具有非常好的泛化能力和环境适应性,并且能够使用几何特征来提高组合效率,具有更可靠的认知心理学依据。此外,几何证据收集过程简洁明了及具有普适性的特点使其表现出极大的应用前景。实验结果证明该方法在应对环境变化、不变性识别、精确指出物体的几何构造、搜索效率与计算量等方面表现出显著的优势。这一尝试有助于理清发生在物体识别过程中的一些带有普遍性的加工环节。

关键词: 模板法,物体检测,证据积累推理

Abstract: Artificial objects usually have very stable shape feature,which has persistent and stable properties in geometry and provides evidence for object recognition.Besides,shape feature is more stable and more discriminative than appearance feature,color feature,gray feature,and gradient feature.The difficulty of object recognition based on shape feature is that objects may change in color,lighting,size,position,pose,and background interference.And we are unable to predict all possible conditions.The variety of objects and conditions make object recognition based on geometric features be a very challenging problem.This paper gave a method based on shape templates,which performs geometric evidence selection,collection,and combination discrimination for the edge segments of images,to find out the target object accurately from background,and it is able to point out the semantic attribute for each line segment of the target object.In essence,the method is solving a global optimal combinatorial optimization problem.Although the complexity of the global optimal combinatorial optimization problem seems to be very high,it is no need to define the complex feature vector and there’s no need for a high price training process.It has very good generalization ability,environmental adaptability,and more solid basis of cognitive psychology.The geometric evidence collection process,which is simple and universal,shows a great application prospect for this method.The experimental results prove that the method shows great advantages in response to the changes in the environment,invariant recognition,pinpointing the geometry of objects,search efficiency,calculation,and etc.This attempt contributes to understanding some universal processing during the process of object recognition.

Key words: Template-based method,Object detection,Evidence accumulation reasoning

唐福宇,危辉. 基于几何证据收集的物体检测[J]. 计算机科学, 2016, 43(10): 304-311. https://doi.org/10.11896/j.issn.1002-137X.2016.10.057

TANG Fu-yu and WEI Hui. Object Detection Based on Geometric Evidence Collecting[J]. Computer Science, 2016, 43(10): 304-311. https://doi.org/10.11896/j.issn.1002-137X.2016.10.057

参考文献

[1] Eysenck M W,Keane M T.Cognitive Psychology:A Student’s Handbook(4th edn)[M].Hove,UK:Psychology Press,2000
[2] Biederman I.Recognition-by-Components:A theory of humanimage understanding[J].Psychological Review,1987,94(1):115-147
[3] Hubel D H,Wiesel T N.Receptive fields of single neurones in the cat’s striate cortex[J].Journal of Physiology,1959,148(3):574-591
[4] Weinschenk S M.100 Things Every Designer Needs to Know About People[M].Pearson Education,2011
[5] Palmer S,Rosch E,Chase P.Canonical perspective and the perception of objects[M].Attention & Performance,IX Hillsdale Nj Erlbaum,1981
[6] Sanocki T,Bowyer K W,Heath M D,et al.Are edges sufficient for object recognition?[J].Journal of Experimental Psychology Human Perception and Performance,1998,24(1):340-349
[7] Berg A C,Berg T L,Malik J.Shape matching and object recognition using low distortion correspondence[C]∥Proceedings of the Computer Vision and Pattern Recognition.2005,1:26-33
[8] Belongie S,Malik J,Puzicha J.Shape matching and object recognition using shape contexts[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(4):509-522
[9] Zhang J,Zhang X,Krim H,et al.Object representation and re-cognition in shape spaces[J].Pattern Recognition.2003,36(5):1143-1154
[10] Ansari N,Delp E J.Partial shape recognition:A landmark-based approach[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1990,12(5):470-483
[11] Marszalek M,Schmid C.Accurate object localization with shape masks[C]∥Proceedings of the Computer Vision and Pattern Recognition.IEEE,2007:1-8
[12] Marszalek M,Schmid C.Accurate object recognition with shape masks[J].International Journal of Computer Vision,2012,97(2):191-209
[13] Mikolajczyk K E A,Mikolajczyk K,Zisserman A,et al.Shape recognition with edge-based features[C]∥Proceedings of the British Machine Vision Conference,2003,2:779-788
[14] Diplaros A,Gevers T,Patras I.Combining color and shape information for illumination-viewpoint invariant object recognition[J].IEEE Transactions on Image Processing,2006,15(1):1-11
[15] Gevers T,Smeulders A W M.Color based object recognition[J].Lecture Notes in Computer Science,1997,32(98):453-464
[16] Martin D R,Fowlkes C C,Malik J.Learning to detect natural image boundaries using local brightness,color,and texture cues[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(5):530-549
[17] Gevers T,Stokman H.Robust Histogram Construction fromColor Invariants for Object Recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(1):113-117
[18] Tsin Y,Collins R T,Ramesh V,et al.Bayesian Color Constancy for Outdoor Object Recognition[C]∥Proceedings of the Computer Vision and Pattern Recognition.IEEE Computer Society,2001:1132
[19] Muselet D,Macaire L.Combining color and spatial information for object recognition across illumination changes[J].Pattern Recognition Letters,2007,28(10):1176-1185
[20] Serre T,Wolf L,Bileschi S,et al.Robust object recognition with cortex-like mechanisms[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(3):411-426
[21] Drimus A,Petersen M B,Bilberg A.Object texture recognition by dynamic tactile sensing using active exploration[C]∥Proceedings of the Robot ad Human Interactive Communication,2012 IEEE.IEEE,2012:277-283
[22] Drimbarean A,Whelan P F.Experiments in colour texture analy-sis[J].Pattern Recognition Letters,2001,22(10):1161-1167
[23] Lazebnik S,Schmid C,Ponce J.A maximum entropy framework for part-based texture and object recognition[C]∥Tenth IEEE International Conference on Computer Vision.IEEE,2005:832-838
[24] Dalal N,Triggs B.Histograms of Oriented Gradients for Human Detection[C]∥Proceedings of the Computer Vision and Pattern Recognition.IEEE Computer Society,2005:886-893
[25] Swain M,Ballard D.Color indexing[J].International Journal of Computer Vision,1991,7(1):11-32
[26] Schiele B,Crowley J L.Recognition without correspondence using multidimensional receptive field histograms[J].Internatio-nal Journal of Computer Vision,2000,36(1):31-50
[27] Linde O,Lindeberg T.Object recognition using composed receptive field histograms of higher dimensionality[C]∥International Conference on Pattern Recognition.Cambridge,U.K.II:2004:1-6
[28] Linde O,Lindeberg T.Composed complex-cue histograms:An investigation of the information content in receptive field based image descriptors for object recognition[J].Computer Vision and Image Understanding,2012,116(4):538-560
[29] Arbelaez P,Maire M,Fowlkes C,et al.From contours to regions:An empirical evaluation[C]∥Proceedings of the Compu-ter Vision and Pattern Recognition.2009:2294-2301
[30] Kovesi P.MATLAB and Octave Functions for Computer Vision and Image Processing.http://www.csse.uwa.edu.au/~pk/research/matlabfns/#edgelink
[31] Wang X,Bai X,Ma T,et al.Fan shape model for object detection[C]∥Proceedings of the Compu-ter Vision and Pattern Recognition.IEEE,2012:151-158
[32] Chia Y S,Rahardja S,Rajan D,et al.Object recognition by discriminative combinations of line segments and ellipses[C]∥Proceedings of the Computer Vision and Pattern Recognition.IEEE,2010:2225-2232
[33] Chia A Y,Rajan D,Leung M K,et al.Object recognition by discriminative combinations of line segments,ellipses and appea-rance features[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,34(9):1758-1772
[34] Shotton J,Blake A,Cipolla R.Contour-based learning for object detection[C]∥Proceedings of the Tenth IEEE International Conference on Computer Vision.2005,1:503-510
[35] Fergus R,Perona P,Zisserman A.Object class recognition by unsupervised scale-invariant learning[C]∥Proceedings of the Computer Vision and Pattern Recognition.2003:264-271
[36] Agarwal S,Roth D.Learning a sparse representation for object detection[C]∥Proceedings of the European Conference on Computer Vision.2002:113-130
[37] Leibe B,Leonardis A,Schiele B.Combined object categorization and segmentation with an implicit shape model[J].Eccv Workshop on Statistical Learning in Computer Vision,2004:17-32
[38] Winn J.The layout consistent random field for recognizing and segmenting partially occluded objects[C]∥Proceedings of the Computer Vision and Pattern Recognition.2006,1:37-44

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed