Flexible Object Recognition in Cluttered Scenes Using Relative Point Distribution Models

Bouganis, Alexandros; Shanahan, Murray

doi:10.1109/icpr.2008.4761712

Cited by 8 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Object detection and localization are critical enabling technologies for robotic assembly automation. Early approaches relied primarily on shape-based methods (Žunić et al , 2010; Bao et al , 2005; Liu et al , 2019; Bohg and Kragic, 2010; Bouganis and Shanahan, 2008). These methods could achieve fast and precise localization given clear object silhouettes by matching candidate shapes to known object templates.…”

Section: Related Workmentioning

confidence: 99%

An MS-TCN based spatiotemporal model with three-axis tactile for enhancing flexible printed circuit assembly

Kang,

Cui,

Wang

et al. 2024

RIA

View full text Add to dashboard Cite

Purpose Current flexible printed circuit (FPC) assembly relies heavily on manual labor, limiting capacity and increasing costs. Small FPC size makes automation challenging as terminals can be visually occluded. The purpose of this study is to use 3D tactile sensing to mimic human manual mating skills for enabling sensing offset between FPC terminals (FPC-t) and FPC mating slots (FPC-s) under visual occlusion. Design/methodology/approach The proposed model has three stages: spatial encoding, offset estimation and action strategy. The spatial encoder maps sparse 3D tactile data into a compact 1D feature capturing valid spatial assembly information to enable temporal processing. To compensate for low sensor resolution, consecutive spatial features are input to a multistage temporal convolutional network which estimates alignment offsets. The robot then performs alignment or mating actions based on the estimated offsets. Findings Experiments are conducted on a Redmi Note 4 smartphone assembly platform. Compared to other models, the proposed approach achieves superior offset estimation. Within limited trials, it successfully assembles FPCs under visual occlusion using three-axis tactile sensing. Originality/value A spatial encoder is designed to encode three-axis tactile data into feature maps, overcoming multistage temporal convolution network’s (MS-TCN) inability to directly process such input. Modifying the output to estimate assembly offsets with related motion semantics overcame MS-TCN’s segmentation points output, unable to meet assembly monitoring needs. Training and testing the improved MS-TCN on an FPC data set demonstrated accurate monitoring of the full process. An assembly platform verified performance on automated FPC assembly.

show abstract

Section: Related Workmentioning

confidence: 99%

An MS-TCN based spatiotemporal model with three-axis tactile for enhancing flexible printed circuit assembly

Kang,

Cui,

Wang

et al. 2024

RIA

View full text Add to dashboard Cite

show abstract

“…Elidan used pairwise spatial relations between landmark points (Elidan et al, 2006). Alexandros presented a method using relative point distribution models for object detection (Bouganis and Shanahan, 2008). These point corresponding approaches are always limited to a localized region around the point.…”

Section: Related Workmentioning

confidence: 99%

“…Many categories are characterized by their shape rather than by color and texture. Although several approaches based on contour features have been proposed (Elidan et al, 2006;Bouganis and Shanahan, 2008;ChengEn and Ling, 2011; The current issue and full text archive of this journal is available at www.emeraldinsight.com/0260-2288.htm The work related to the paper is partly funded by the Natural Science Fund of Higher Education of Jiangsu Province (11KJB510024) and partly funded by Natural Science Foundation of China, under contract number 61105098. The authors would like to thank the anonymous reviewers and the editors for their many helpful suggestions.…”

Section: Introductionmentioning

confidence: 99%

A learning algorithm for model‐based object detection

et al. 2013

View full text Add to dashboard Cite

Purpose -Detecting objects in images and videos is a difficult task that has challenged the field of computer vision. Most of the algorithms for object detection are sensitive to background clutter and occlusion, and cannot localize the edge of the object. An object's shape is typically the most discriminative cue for its recognition by humans. The purpose of this paper is to introduce a model-based object detection method which uses only shape-fragment features. Design/methodology/approach -The object shape model is learned from a small set of training images and all object models are composed of shape fragments. The model of the object is in multi-scales. Findings -The major contributions of this paper are the application of learned shape fragments-based model for object detection in complex environment and a novel two-stage object detection framework. Originality/value -The results presented in this paper are competitive with other state-of-the-art object detection methods.

show abstract

“…Similar to other fields of machine learning, the success in performing many tasks in computer vision can be subject to, and highly dependent on, the goodness of the features. Hence, due to the ability of image descriptors to detect and extract useful information, i.e., keypoints and features, they have been widely deployed in computer vision as a prior step in object detection [20], [21], object recognition [22], image classification [23], image registration [24], and image segmentation [25]. Some typical examples of such descriptors are Haralick texture features [10], Local Binary Pattern (LBP) [11], Scale-Invariant Feature Transform (SIFT) [12], Speeded-Up Robust Features (SURF) [13], Fast Retina Keypoint (FREAK) [17], and KAZE features [18].…”

Section: Introductionmentioning

confidence: 99%

Automatically Evolving Rotation-Invariant Texture Image Descriptors by Genetic Programming

Al-Sahaf¹,

Al-Sahaf²,

Xue³

et al. 2020

Preprint

View full text Add to dashboard Cite

© 2016 IEEE. In computer vision, training a model that performs classification effectively is highly dependent on the extracted features, and the number of training instances. Conventionally, feature detection and extraction are performed by a domain expert who, in many cases, is expensive to employ and hard to find. Therefore, image descriptors have emerged to automate these tasks. However, designing an image descriptor still requires domain-expert intervention. Moreover, the majority of machine learning algorithms require a large number of training examples to perform well. However, labeled data is not always available or easy to acquire, and dealing with a large dataset can dramatically slow down the training process. In this paper, we propose a novel genetic programming-based method that automatically synthesises a descriptor using only two training instances per class. The proposed method combines arithmetic operators to evolve a model that takes an image and generates a feature vector. The performance of the proposed method is assessed using six datasets for texture classification with different degrees of rotation and is compared with seven domain-expert designed descriptors. The results show that the proposed method is robust to rotation and has significantly outperformed, or achieved a comparable performance to, the baseline methods.

show abstract

Flexible Object Recognition in Cluttered Scenes Using Relative Point Distribution Models

Cited by 8 publications

References 10 publications

An MS-TCN based spatiotemporal model with three-axis tactile for enhancing flexible printed circuit assembly

An MS-TCN based spatiotemporal model with three-axis tactile for enhancing flexible printed circuit assembly

A learning algorithm for model‐based object detection

Automatically Evolving Rotation-Invariant Texture Image Descriptors by Genetic Programming

Contact Info

Product

Resources

About