Image moments: generic descriptors for decoupled image-based visual servo

Tahri, Omar; Chaumette, François

doi:10.1109/robot.2004.1307985

Cited by 23 publications

(34 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Even if successful, all the required markers may not be visible in the field of view. Recent work, [21,22] has shown that image moments can be used as features for servoing, Avoiding the need for markers placed in the scene. We define the servoing task as bringing the root tip close to the center of the field of view.…”

Section: Image Processing For Visual Servoingmentioning

confidence: 99%

A high throughput robot system for machine vision based plant phenotype studies

Subramanian

Spalding

Ferrier

2012

Machine Vision and Applications

View full text Add to dashboard Cite

This work demonstrates how a high throughput robotic machine vision systems can quantify seedling development with high spatial and temporal resolution.The throughput that the system provides is high enough to match the needs of functional genomics research. Analyzing images of plant seedlings growing and responding to stimuli is a proven approach to finding the effects of an affected gene. However, with 10 4 genes in a typical plant genome, comprehensive studies will require high throughput methodologies. To increase throughput without sacrificing spatial or temporal resolution, a 3 axis robotic gantry system utilizing visual servoing was developed. The gantry consists of direct drive linear servo motors that can move the cameras at a speed of 1 m/s with an accuracy of 1 µm, and a repeatability of 0.1 µm. Perpendicular to the optical axis of the cameras was a 1 m 2 sample fixture holds 36 Petri plates in which 144 Arabidopsis thaliana seedlings (4 per Petri plate) grew vertically along the surface of an agar gel. A probabilistic image analysis algorithm was used to locate the root of seedlings and a normalized gray scale variance measure was used to achieve focus by servoing along the optical axis. Rotation of the sample holder induced a gravitropic bending response in the roots, which are approximately 45 µm wide and several millimeter in length. The custom hardware and software described here accurately quantified the gravitropic responses of the seedlings in parallel at approximately 3 min intervals over an 8-h period. Here we present an overview of our system and describe some of the necessary capabilities and challenges to automating plant phenotype studies.

show abstract

Section: Image Processing For Visual Servoingmentioning

confidence: 99%

A high throughput robot system for machine vision based plant phenotype studies

Subramanian

Spalding

Ferrier

2012

Machine Vision and Applications

View full text Add to dashboard Cite

show abstract

“…In order to have a better and easier control over the robot trajectory, approximatively decoupled features are chosen as proposed in [3], [16]. Six visual features derived from the moments of the set of points have been chosen to control the six DOF of the robot.…”

Section: A Main Task Function Using Visual Servoingmentioning

confidence: 99%

“…The two first features are the position x g and y g of the center of gravity. Since the considered object is discrete, the third feature a is based on the centered moments of order 2 to control the range between the robot and the target, as proposed in [16]. The fourth feature α mainly rotates the camera around the optical axis.…”

Section: A Main Task Function Using Visual Servoingmentioning

confidence: 99%

A new redundancy formalism for avoidance in visual servoing

Mansard

Chaumette

2005

2005 IEEE/RSJ International Conference on Intelligent Robots and Systems

Self Cite

View full text Add to dashboard Cite

To cite this version:Abstract-The paper presents a new approach to construct a control law that realizes a main task and simultaneously takes supplementary constraints into account. Classicaly, this is done by using the redundancy formalism. If the main task does not constrain all the motions of the robot, a secondary task can be achieved by using only the remaining degrees of freedom (DOF). We propose a new general method that frees up some of the DOF constrained by the main task in addition of the remaining DOF. The general idea is to enable the motions produced by the secondary control law that help the main task to be completed faster. The main advantage is to enhance the performance of the secondary task by enlarging the number of available DOF. In a formal framework, a projection operator is built which ensures that the secondary control law does not disturb the main task. A control law can be then easily computed from the two tasks considered. Experiments that implement and validate this approach are proposed. The visual servoing framework is used to position a 6-DOF robot while simultaneously avoiding occlusions and joint limits.

show abstract

“…Problems (18) and (17) have been solved through the function FMINSEARCH of MATLAB by evaluating α(C) andρ(C) as described in Section III-C. Problem (17) has been initialized with the solution of problem (18), and problem (18) has been initialized with C = 0 3×(δ−1) . Figure 1a shows the current and desired locations denoted by F and F * , and the observed features corresponding to the biggest white circles.…”

Section: Numerical Examplesmentioning

confidence: 99%

“…In partitioning methods, the degrees of freedom of the camera are grouped according to some criteria and differently controlled (see for example [9]- [11]). Other methods, such as visual servoing based on navigation functions [12], path-planning techniques [13]- [15], visual servoing invariant to the intrinsic parameters [16], visual servoing based on image moments [17] and methods based on the generation of circular-like trajectories [18], have been proposed in the literature.…”

Section: Introductionmentioning

confidence: 99%