Verification of illumination tolerance for photo-model-based cloth recognition

Phyu, Khaing Win; Funakubo, Ryuki; Hagiwara, Ryota; Tian, Haishan; Minami, Mamoru

doi:10.1007/s10015-017-0391-0

Cited by 6 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The robot shown in Fig. 1 has been confirmed to be able to identify 12 different deformable clothes, estimate the pose and handle them under illumination varieties (Phyu et al, 2018a and2018b). Previous studies (Phyu et al, 2017 andTian, Kou, Li andMinami, Journal of Advanced Mechanical Design, Systems, andManufacturing, Vol.14, No.4 (2020) 2018c) have also verified that the photo-model-based recognition method can detect the partly hidden cloth.…”

Section: Contributionsmentioning

confidence: 63%

“…Since the photo-model can be made at once by taking a photo of the target object, the target can be recognized instantly, and the pose can be estimated immediately. Figure 1 shows the photo-model-based handling robot (pick and place) introduced by our previous researches (Funakubo et al, 2016 andPhyu et al, 2016 and). The proposed system aims at picking up clothes after a robot recognizes it and classifies the clothes into a collection box.…”

Section: Contributionsmentioning

confidence: 99%

See 1 more Smart Citation

Real-time pose tracking of 3D targets by photo-model-based stereo-vision

Tian

Kou

et al. 2020

JAMDSM

Self Cite

View full text Add to dashboard Cite

Nowadays, robots have been utilized to perform a wide variety of tasks instead of humans, in which those equipped with hand-eye cameras have been proved to be useful in factories for handling metal parts. However, there have been difficulties for robots with vision sensors like cameras to detect a target objects' 3D pose accurately, especially in the case that the target object is unique and the shape is arbitrary, and further, the object is moving. Visual servoing to moving target may enable the robot to pursue an animal to catch it. Aiming at achieving identification ability of a unique object, this paper utilized our previous study concerning photo-model-based object recognition. Based on the study, this paper applies the photo-model-based pose estimation method to video-rate stereo-vision position and orientation (pose) tracking, and it has been applied to an arbitrary-shaped moving target. To confirm the abilities of the proposed stereo-vision-based method to pursue an arbitrary target, real-time 3D pose estimation experiment to track the relative pose between a moving target and hand-eye robot have been conducted. The realtime experimental results show the proposed system can recognize the designated arbitrary-shaped target and track its pose in real-time.

show abstract

Section: Contributionsmentioning

confidence: 63%

Section: Contributionsmentioning

confidence: 99%

Real-time pose tracking of 3D targets by photo-model-based stereo-vision

Tian

Kou

et al. 2020

JAMDSM

Self Cite

View full text Add to dashboard Cite

show abstract

“…Furthermore, in handling different lighting conditions, the practicality of the photo-model-based method was tested experimentally [ 28 ]. The experiments focused on two common light sources: fluorescent and light-emitting diode (LED) lighting.…”

Section: Related Workmentioning

confidence: 99%

A Variable Photo-Model Method for Object Pose and Size Estimation with Stereo Vision in a Complex Home Scene

Tian

Wang

2023

Sensors

Self Cite

View full text Add to dashboard Cite

Model-based stereo vision methods can estimate the 6D poses of rigid objects. They can help robots to achieve a target grip in complex home environments. This study presents a novel approach, called the variable photo-model method, to estimate the pose and size of an unknown object using a single photo of the same category. By employing a pre-trained You Only Look Once (YOLO) v4 weight for object detection and 2D model generation in the photo, the method converts the segmented 2D photo-model into 3D flat photo-models assuming different sizes and poses. Through perspective projection and model matching, the method finds the best match between the model and the actual object in the captured stereo images. The matching fitness function is optimized using a genetic algorithm (GA). Unlike data-driven approaches, this approach does not require multiple photos or pre-training time for single object pose recognition, making it more versatile. Indoor experiments demonstrate the effectiveness of the variable photo-model method in estimating the pose and size of the target objects within the same class. The findings of this study have practical implications for object detection prior to robotic grasping, particularly due to its ease of application and the limited data required.

show abstract

“…Three different illumination conditions (100 lx, 700 lx, and 1300 lx) using fluorescent light and light-emitting diode (LED) separately were given as experimental environments in (Funakubo R et al, 2017). In (Phyu, Khaing Win, et al, 2017), five different illumination conditions (100 lx, 400 lx, 700 lx, 1000 lx and 1300 lx) using fluorescent light and light-emitting diode (LED) were separately simulated for experiments. Fluorescent light was used for the experiments of the present paper.…”

Section: Robustness Confirmation Against Illumination and Lighting Somentioning

confidence: 99%

“…In our previous study, two different light sources, fluorescent light and light-emitting diode (LED), were used to provide with different illuminations environments. Why these illumination conditions were chosen has been discussed in (Phyu, Khaing Win, et al, 2017). According to the experimental results, the recognition performances under fluorescent light are better than light-emitting diode.…”

mentioning

confidence: 99%

Verification of photo-model-based pose estimation and handling of unique clothes under illumination varieties

Phyu

Funakubo

Hagiwara

et al. 2018

JAMDSM

Self Cite

View full text Add to dashboard Cite

Human can recognize and handle (pick and place) easily the objects with a variety of different shapes, colors, sizes, and humans' eyes are adaptable to various light environments with a certain tolerance. However, it is difficult for robots to recognize deformable objects such as cloth, string, etc., especially if an object is unique. Additionally, there have been difficulties for robots with vision sensors (cameras) to accurately detect and handle objects under various light environments. This paper proposes a cloth handling system that recognizes an unique cloth appeared in front of a robot by a photo-model-based approach. The photo-model-based approach has been adopted since the photo-model can be made at once by taking a photo of the unique cloth. In proposed cloths' pose estimation method, a photo-model projected from 3D to 2D is used, where this system does not need defining the object's size, shape, design, color and weight. It detects the cloth through model-based matching method and Genetic Algorithm (GA). The handling performance by the proposed method with dual-eyes cameras has been verified, revealing that the proposed system has leeway to recognize and handle the unique cloth in lighting varieties from 100 lx to 1300 lx. In addition, 3D recognition and handling accuracy have been confirmed to be practically effective by conducting the recognition/handling experiments under different light conditions.Keywords : Photo-model-based cloth recognition, Handling, Visual servoing, Genetic Algorithm, Dual-eyes cameras, Illumination IntroductionFrom the instant of birth, human beings are thought to be talented at managing their activities under such variability as climates, light environments, temperatures, etc. While human beings can conduct intended tasks in pending circumstances, an automated robot is not adept at being similarly adaptable. Therefore, the researchers have tried to improve the abilities of automated robots.Nowadays, industrial robots have been utilized to perform a wide variety of tasks instead of human workers. These automated robots are required to handle a wide variety of deformable objects including cloths, strings, ropes, electric cables and so on. Of course, handling deformable objects is difficult than handling rigid objects. A robot control technology using visual information, called as visual servoing, has been playing an important role in the applications where deformable things are recognized and handled by a robot.Each item of the deformable target objects has various possibilities of the poses (positions and orientations) to be recognized and handled, requiring ability with respect to both vision-based recognition and visual servoing. In (MaitinShepard J et al., 2010), the cloth-grasping points are detected using four cameras without using other sensors for a towel folding application by robots. The main task in (Maitin-Shepard J et al., 2010) is to detect the corners of cloth instead of recognition a whole cloth. The recognition of cloth shape based on strategic observ...

show abstract

Verification of illumination tolerance for photo-model-based cloth recognition

Cited by 6 publications

References 9 publications

Real-time pose tracking of 3D targets by photo-model-based stereo-vision

Real-time pose tracking of 3D targets by photo-model-based stereo-vision

A Variable Photo-Model Method for Object Pose and Size Estimation with Stereo Vision in a Complex Home Scene

Verification of photo-model-based pose estimation and handling of unique clothes under illumination varieties

Contact Info

Product

Resources

About