Description of shape patterns using circular arcs for object detection

Chang, Wonil; Lee, Sang Yup

doi:10.1049/iet-cvi.2011.0180

Cited by 10 publications

(5 citation statements)

References 50 publications

(110 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this step query object i.e. image of object to be detected is converted from RGB to HSL color space using equations (1) to (9). The main reason to convert RGB to HSL is that in HSL color model L (lightness) represents change in color due to change in brightness, so if we neglect the effect of brightness on color any color can be represented by only two values i.e.…”

Section: A Formation Of Black and White Clustermentioning

confidence: 99%

“…Shape based information is provided by edges and corners of segmented object. Shape based information is used by many authors in different way such as Chang and Lee [1] used segments of circular arcs to describe the shape of object while Gonzalez-Sosa et al [2] used shape-based feature approaches, such as shape contexts and contour coordinates to recognize person from his millimeter waves images. To extract proper shape information, it is very important that object should be segmented properly.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Object Detection In Cluttered Background using Color Clusters

Gode,

Khobragade

2019

IJEAT

View full text Add to dashboard Cite

Object detection in presence of complex background and illumination variation is important image analysis problem with many applications. Most of the object detection algorithms use local image descriptors which are computed from interest points based on luminance information and neglect precious color information of an object. If appearances of the object to be detected contain multiple colors in non-homogeneous distributions then it makes it difficult to detect these objects using shape features. In this context, we propose a robust algorithm designed to detect a class of objects using a descriptor which is computed from color information of an object. Clusters are formed in Hue and Saturation (HS) color space of an object using k-means clustering and cluster analysis based on number of pixels belong to each cluster, object detection is performed. Use of clustering algorithm in color space of an object to form descriptor reduces the large dimensionality of the histogram bins in the computation. The performance of the algorithm is demonstrated by experimentation carried out on standard dataset GroZi-120. Experimental results shows that the proposed algorithm is insensitive to scaling, object rotation, illumination variations and capable of handling cluttered background effectively. Finally results shows that proposed algorithm outperforms closely related algorithm by a decisive margin.

show abstract

Section: A Formation Of Black and White Clustermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Object Detection In Cluttered Background using Color Clusters

Gode,

Khobragade

2019

IJEAT

View full text Add to dashboard Cite

show abstract

“…A first class of approaches is the countingby-detection methods [41], which formulate the problem as a detection task. Typical solutions rely on local features, such as histogram of oriented gradients (HOG) [9,12], local binary patters [8], or shape [22]. Nevertheless, leaf detection is a challenging task, since leaf surface is almost featureless and shape information is unreliable under heavy occlusion, as it can be seen in Figure 1.…”

Section: Related Workmentioning

confidence: 99%

Learning to Count Leaves in Rosette Plants

Giuffrida¹,

Minervini²,

Tsaftaris³

2015

Procedings of the Proceedings of the Computer Vision Problems in Plant Phenotyping Workshop 2015

121

View full text Add to dashboard Cite

Counting the number of leaves in plants is important for plant phenotyping, since it can be used to assess plant growth stages. We propose a learning-based approach for counting leaves in rosette (model) plants. We relate image-based descriptors learned in an unsupervised fashion to leaf counts using a supervised regression model. To take advantage of the circular and coplanar arrangement of leaves and also to introduce scale and rotation invariance, we learn features in a log-polar representation. Image patches extracted in this log-polar domain are provided to K-means, which builds a codebook in a unsupervised manner. Feature codes are obtained by projecting patches on the codebook using the triangle encoding, introducing both sparsity and specifically designed representation. A global, per-plant image descriptor is obtained by pooling local features in specific regions of the image. Finally, we provide the global descriptors to a support vector regression framework to estimate the number of leaves in a plant. We evaluate our method on datasets of the Leaf Counting Challenge (LCC), containing images of Arabidopsis and tobacco plants. Experimental results show that on average we reduce absolute counting error by 40% w.r.t. the winner of the 2014 edition of the challenge -a counting via segmentation method. When compared to state-of-the-art density-based approaches to counting, on Arabidopsis image data ∼75% less counting errors are observed. Our findings suggest that it is possible to treat leaf counting as a regression problem, requiring as input only the total leaf count per training image.

show abstract

“…The first need is a robust feature set that allows the human form to be discriminated cleanly, even in cluttered backgrounds under difficult illumination. Colour characteristics, shape, texture, gradient features are the common choices (Nedevschi, 2009;Chang, 2013). Histograms of oriented gradients (HOG) introduced in Dalal (2005) is currently one of the most performant and widely used methods for visual people detection in RGB data.…”

Section: Introductionmentioning

confidence: 99%

Multi-person detecting and tracking based on RGB-D sensor for a robot vision system

Jiang

Wang

Qiu

2017

IJES

View full text Add to dashboard Cite

Abstract:In this paper, we address the problem of automatically detecting and tracking a variable number of objects in complex scenes using a RGB-D sensor on the robot system. We propose a novel approach for multi-object detecting by fusing RGB information and depth information. Meanwhile, this paper presents a robust multi-cue approach for multi-object tracking. A spatiotemporal object representation is proposed, which combines a generative colour model and a discriminative texture classifier. We employ a Bayesian framework based on particle filtering to achieve integrated object detection and tracking from a robot vision system. The experimental results show that the proposed method yields good tracking performance in real world environment.Keywords: combined multi-object detection; multi-cue object representation; particle filtering; RGB-D sensor. Dusheng Wang is a Professor of Electrical Information Engineering at Dalian Nationalities University. His research includes multimedia signal processing, speech coding and computer graphics. He was involved in more than 30 projects since 1983, including in charge of the project 'Research on low bit rate speech coding based on nonlinear time-frequency interpolation' sponsored by National Natural Science Foundation of China. He was employed as a research fellow and a scientist in Nanyang Technological University and Institute for Infocomm Research, Singapore to work with the projects in computer graphics. He developed various practical speech coders applied in wireless channel, business and military communication systems. He was awarded to over ten projects and has over 30 refereed technical publications.Tianshuang Qiu received his BS from Tianjin University, Tianjin, China, the MS from the Dalian University of Technology, Dalian, China and PhD from Southeastern University, Nanjing, China, all in Electrical Engineering, in 1983, 1993, and 1996. He was a post-doctoral researcher with the Department of Electrical Engineering, Northern Illinois University, DeKalb. Multi-person detecting and tracking based on RGB-D sensor for a robot vision system 55He is currently a Professor with the Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology. His current research interests include non-Gaussian and non-stationary signal processing, radio frequency signal processing, and image processing.

show abstract

Description of shape patterns using circular arcs for object detection

Cited by 10 publications

References 50 publications

Object Detection In Cluttered Background using Color Clusters

Object Detection In Cluttered Background using Color Clusters

Learning to Count Leaves in Rosette Plants

Multi-person detecting and tracking based on RGB-D sensor for a robot vision system

Contact Info

Product

Resources

About