Search results

The purpose of this paper is to present classification schemes for the crystallization state of proteins utilizing image processing.

Two classification schemes shown here are combined sequentially.

The correct ratio of experimental result using the method presented here is approximately 70 per cent.

The paper is a contribution to automated evaluation crystal growth, combining two classifiers based on specific visual feature, sequentially.

The purpose of this paper is to propose an imaging process method for automatically extracting assessment target regions in images of crystallization wells.

The proposed method detects the target by image processing based on saturation variance in images of crystallization wells.

The proposed method shows high accurate detection performance in a short time.

The method is applicable to automatic crystallization machine. Especially, it is expected to implement to TERA system of RIKEN.

The paper demonstrates target area detection from microscope images based on saturation variance in images of crystallization wells.

The purpose of this paper is to propose a novel wireless sensor node (SN), with impact resistant capability, for launch deployment into closed areas. In disaster areas, gathering disaster area information is crucially important to prevent secondary disasters. However, gathering information is usually associated with the risk of death and/or accident for rescue workers in closed areas. The authors propose the SN for gathering information in dangerous places, inaccessible to rescue workers and robots, by utilizing launch deployment.

Buffer material is essential when designing an impact‐resistant structure. The authors adopted the air cushion as general buffer material when considering the directional characteristics of sensor mounting and wireless communication quality and developed the expression for determining the thickness of the air cushion using the parameters of SN size, mass, air pressure and acceleration. The authors developed a sensor node with impact resistant structure by utilizing the proposed determination method of air cushion thickness.

In the evaluation of impact resistant structure in free fall, launch deployment, the authors verified that the impact resistant structure protected the SN, and the performance of configured devices on the SN. Then, the authors examined the effect of the impact‐resistant structure on wireless communication between SNs. The structure had no effect on electric field intensity, throughput, or packet jitter, which confirmed that the wireless communication capacity was unaffected by the structure.

In this paper, a new design method is stated for a sensor node with an impact‐resistant structure by utilizing an air cushion as a general buffer material.

The purpose of this paper is to develop a new wheel control scheme for wheeled vehicle with passive linkage mechanism which realizes high step‐overcoming performance.

Developing wheeled vehicle realizes omni‐directional motion on flat floor using special wheels and passes over non‐flat ground using the passive suspension mechanism. The vehicle changes its body shape and wheel control references according to ground condition when it runs over the rough terrain.

Utilizing the proposed wheel control scheme, the slip ratio and the disturbance ratio of the wheel reduce when the vehicle passes over the step and its step‐overcoming performance is improved.

The paper's key idea is modification of its kinematic model referring to the body configuration dynamically and using this model for wheel control of the vehicle. The controller adjusts the wheel control references when the vehicle passes over the rough terrain changing the body shape and reduces the slippage and the rotation error of wheels.

The purpose of this study is to develop a new mobile system using “leg‐type crawler” mechanism. Our proposed “leg‐type crawler” mechanism realizes flexible mobility on rough terrain environments. The mobile system, which consists of four leg‐type crawler units including mechanisms and controller, can adopt several moving configurations. Since we adopt unimodular structure for the leg‐type crawlers, we can replace the legs easily. This paper shows the outline of the proposed leg‐type crawler units, as well as basic experimental results, proposal of basic control law for this system and verification by experiment.

The aim is to develop a force assistance system for standing‐up which prevents the decreasing of physical strength of the patient by using their remaining physical strength.

The system realizes the standing up motion using the support bar with two degrees of freedom and the bed system which can move up and down. For using the remaining physical strength, our system uses the motion pattern which is based on the typical standing up motion by nursing specialist as control reference.

The assistance system realizes the natural standing up motion by nursing specialist and it is effective to assist the aged person to stand up without reducing their muscular strength.

The first idea is distributed system which controls the support bar and the bed system with coordination among them. The second idea is the combination of force and position control.

The purpose of this paper is to propose a state discrimination for crystallization samples (droplets), the purpose of which is to discriminate between diffractable extracts (crystal) and other objects.

The line feature from the image of the protein droplet was extracted and the state discriminated using a classifier based on line features. A support vector machine is used as the classifier.

In order to verify the performance of the proposed method, the growth state was discriminated experimentally using the images taken by TERA, an automated crystallization system. The correction ratio was determined to exceed 80 percent.

Contribution to automated evaluation process of the growth state of protein crystallization samples.

The purpose of this paper is to propose a sensor system and its measuring strategy. The proposed system is regarded as a total system of a robot and a sensor. It can make full use of a robot's degrees of freedom (DOF) for not only locomotion but also sensing.

The sensor system is composed of a connected crawler robot and a simple sensor unit. The connected crawler robot consists of five connected flat crawler stages; each stage has motor‐driven crawlers on its left and right side. The left and right crawlers are driven by motors independently. The five stages are connected by motor‐driven joints, which can be arbitrarily controlled. The sensor unit is made up of a position sensitive detector and two active joints. Generally, all sensors have a measurable range limit. If the object is bigger than the measurable range, it is impossible to recognize the object's shape. However, this sensor system compensates for this problem due to a combination of sensor unit's motion and robot's motion. The robot's DOF are ordinarily used for going forward. In this moving, if the sensor detects a bigger object than its measurable range, the robot's DOF are used for lifting up sensor unit. Therefore, the robot's DOF are exploited for both locomotion and scanning.

The experiment was done by using proposed measurement strategies. Through this experiment, the maximum distance error was 0.037 m, that fulfils the required sensor's accuracy. This experiment confirmed that the measurable range was expanded by making full use of the robot's DOF and sensor's DOF. The robot's DOF which is ordinarily used for moving is applicable to expand measurable range. Therefore, it was possible to expand measurable range by making full use of the robot's DOF without mounting extra equipments or mechanism on the robot. Hence, it was enable to exert enough sensing function even if the sensor's structures and robot's structure are also simple, by using both the robot and the sensor's functions effectively and simultaneously.

General existing mobile robot systems have both mobile function and sensing function, they are completely separate. However, this sensor system is regarded as a total system of robot and sensor. It can make full use of a robot's DOF for not only locomotion but also sensing. This sensor system makes full use of a robot's DOF. Hence, it is enable to exert enough sensing function even if the sensor's structure and robot's structure are also simple. There is originality on this point.

The purpose of this paper is to propose an automatic interpolation method for binarized confocal laser scanning microscopy (CLSM) images of a premotor neuron in the silkworm moth.

Partial deficiencies occur in binary images through the form extraction process because of noises in a CLSM image series. The proposed method selects several points from a binarized image series and connects these points with a Bezier curve based on premotor neuron characteristics in order to interpolate partial deficiencies.

To verify the availability of the proposed method, a three‐dimensional form of a premotor neuron of a silkworm moth was extracted. The results of each branch's relation of connection and of the interpolated neuron thickness show that the proposed method realizes to interpolate partial deficiencies and to extract three‐dimensional form of the premotor neuron.

The proposed method contributes to realize efficient premotor extraction process using image‐processing techniques. The extracted result by proposed method can be utilized for the form comparison among many data of the premotor neurons quickly. Moreover, it also contributes to provide the parameters of an accurate neuron model for realizing computer simulation of electrical of the neurons.

The proposed method extracts not only a topological form but also a premotor neuron's thickness by interpolating partial deficiencies based on specific characteristics of the neuron. Thickness values of the neuron are an important factor for a simulating accurate electrical response of the neuronal circuit.