The purpose of this study is to develop a crisp nonlinear method of programming to address production inventory problems based on both production inventory and conditions. In addition, we use a statistical confidence interval to derive level (1-β, 1-α) interval-valued fuzzy numbers, in order to solve problems in nonlinear programming for production inventory in the fuzzy sense.
Recent advances in 3D have increased the importance of stereoscopic content creation and processing. Therefore, converting existing 2D videos into 3D videos is very important for growing 3D market. The most difficult task in 2D-to-3D video conversion is estimating depth map from single-view frame images. Thus, in this paper, we propose a novel motion-based 2D to 3D video conversion method. The method first determines the motion type using the optical flow estimation. Then, different depth estimation processes are performed based on the motion type. For global motion, the depth from motion parallax provides the final depth map. For local motion, the depth from template together with the bilateral filter is used to produce the depth map. Finally, the left- and right-view images are synthesized to generate realistic stereoscopic results for viewers. During the process, the visual artifacts of the synthesized virtual views are effectively eliminated by recovering the separation and loss of foreground objects. A comparative study and quantitative evaluation with other conversion methods are carried out, which demonstrate that better overall quality results may be obtained using the proposed method.
There are people who cannot distinguish between specific colors easily. This paper presents an improvement to a system for the color vision deficient. The system consists of a camera that acquires an image of an object and a projector that projects light on that object. One of the features of the system is that it handles real-life objects. When objects have colors that cannot be distinguished, the system converts the color to a distinguishable one using the projector. The improvement in the proposed system is that it produces images with patterns and blinking light to handle conventional color conversion operations that cannot produce distinguishable color images because of excessive multiplicity of color combinations. We verify through experiments the effectiveness of the proposed color projection system with its patterns and blinking light.
Level evaluation system in cardiotocography (CTG) taken in the real medical practice is constructed for the use of e-learning materials. The system consists of 3 parts, preprocessing for FHR by particle filter, extraction process of deceleration information, and level evaluation. The extraction and evaluation processes are executed by fuzzy inference. To check the effectiveness of the proposed system, level evaluation experiments are executed for the real CTG data extracted from records at the Kochi Medical School hospital. The experimental results show that the extracted CTG features are suitable for the material, and that the calculated level of the patients with abnormal pH value is evaluated as abnormal situations. The proposed system can provide the possible data for the materials for evaluation learning of real CTG.
Generic object recognition algorithms usually require complex classification models because of intrinsic difficulties arising from problems such as changes in pose, lighting conditions, or partial occlusions. Decision trees present an inexpensive alternative for classification tasks and offer the advantage of being simple to understand. On the other hand, a common scheme for object recognition is given by the appearances of visual words, also known as the bag-of-words method. Although multiple co-occurrences of visual words are more informative regarding visual classes, a comprehensive evaluation of such combinations is unfeasible because it would result in a combinatorial explosion. In this paper, we propose to obtain the multiple co-occurrences of visual words using a variant of the CLIQUE subspace-clustering algorithm for improving the object recognition performance of simple decision trees. Experiments on standard object datasets show that our method improves the accuracy of the classification of generic objects in comparison to traditional decision tree techniques that are similar, in terms of accuracy, to ensemble techniques. In future we plan to evaluate other variants of decision trees, and apply other subspace-clustering algorithms.
This study proposes a vision-based mowing boundary detection algorithm for an autonomous lawn mower. An autonomous lawn mower requires high moving accuracy for efficient mowing. This problem is solved by using a vision system to detect the boundary of two regions, i.e., before and after the lawn mowing process. The mowing boundary cannot be detected directly because it is ambiguous. Therefore, we utilize a texture classification method with a bank of filters for classifying the input image of the lawn field into two regions as mentioned above. The classification is performed by threshold processing based on a chi-squared statistic. Then, the boundary line is detected from the classified regions by using Random sample consensus (RANSAC). Finally, we apply the proposed method to 12 images of the lawn field and verified that the proposed method can detect a mowing boundary line with centimeter accuracy in a dense lawn field.
We propose a discriminative and compact scene descriptor for single-view place recognition that facilitates long-term visual SLAM in familiar, semi-dynamic, and partially changing environments. In contrast to popular bag-of-words scene descriptors, which rely on a library of vector quantized visual features, our proposed scene descriptor is based on a library of raw image data (such as an available visual experience, images shared by other colleague robots, and publicly available image data on the Web) and directly mine it to find visual phrases (VPs) that discriminatively and compactly explain an input query/database image. Our mining approach is motivated by recent success achieved in the field of common pattern discovery – specifically mining of common visual patterns among scenes – and requires only a single library of raw images that can be acquired at different times or on different days. Experimental results show that, although our scene descriptor is significantly more compact than conventional descriptors, its recognition performance is relatively high.
Topic evolution analysis helps to understand how the topics evolve or develop along the timeline. Aiming at the problem that existing researches did not mine the latent semantic information in depth and needed to pre-determine the number of clusters, this paper proposes cluster topic model based method to analyze topic evolution analysis. Firstly, a new topic model, namely cluster topic model, is built to complete document clustering while mining latent semantic information. Secondly, events are detected according to the cluster label of each document and evolution relationship between any two events is identified based on the aspect distributions of documents. Finally, by choosing the representative document of each event, topic evolution graph is constructed to display the development of the topic along the timeline. Experiments are presented to show the performance of our proposed technique. It is found that our proposed technique outperforms the comparable techniques in previous work.
An implicit solution of Colebrook-White equation was used in calculating the friction factor for commercial steel pipes using Newton-Raphson method with Reynolds number ranging from 4.0×103 to 1.3×107. Initial value for iterative friction factor estimation was based on expanded form of Colebrook-White equation for larger values of Reynolds number with tolerance value of 1.0×10-8. Numerical results were compared with known explicit solutions and iterative procedure proposed by Lester in which, their mean difference, root-mean square deviation, mean relative error and correlation coefficient were evaluated. Correlation coefficients equal to unity and overall mean relative error of 4.821×10-8 were achieved for all fifteen (15) pipe cases with nominal diameters ranging from 100 mm to 1,500 mm when compared with iterative solution suggested by Lester. Student’s t-test for paired data was also used which yielded a calculated t-value of -5.406×10-4. Combining the piping network design criteria with the logical structure of friction factor calculation determines the pipe size of distribution network and defines the boundaries of chilled-water velocities at different pressure drop limits as a function of commercial steel pipe diameter according to ANSI B36.1.
This paper presents an aggregation behavior derived from fluid characteristics by adapting Smoothed Particle Hydrodynamics (SPH) Technique. The most basic behavior in a swarm-like system is aggregation. The essential requirement of a swarm is to aggregate or collect itself in proximity to a singular point in order to execute higher level swarm behaviors. The aggregation behavior is further put into use by initiating a near convergence status in a single target enclosing it by the swarm with a given specific distance by using different fluid containers. In this paper, there are three fluid containers each is introduced with different characteristics. These containers are plane, spherical and toroidal containers. Using computer simulations with different trials, the proponents were able to determine the accuracy of containing the swarm elements in a desirable area. Furthermore, the ability of the swarm to maintain collectiveness is tested. The experiment results showed that the plane fluid container yielded an accuracy of 84.88%. A spherical fluid container displayed an accuracy of 95.23%. And using toroidal particle container showed an accuracy of 92.44%.
Swarm robotics is a collection of mobile robots that displays swarm behavior. This paper presents a simulator of slime mold amoeba inspired swarm robot for underwater wireless communication system. The slime mold inspired robotic swarm is used to overcome the challenges of transmitting data in a large underwater environment. Underwater communication systems today are primarily acoustic technology and characterized by limited and distance dependent bandwidth, presence of multipath, and low speed of sound propagation. The robots navigate and seek the shortest path creating a virtual connection between the data transmitter and receiver similar to the foraging behavior of swarms. Each individual robot going back and forth from the transmitter to the receiver and vice-versa acts as a “physical” carrier of the data. Swarm robots navigate using swarm level intelligence based on the signal propagation technique used by slime mold amoeba aggregation using acoustics communication. The robot swarm system is developed, simulated and tested using the coded simulator. Using the slime mold inspired swarm robot system; the simulation successfully performed the data “foraging” scenario and showed the ability of the swarm to provide a virtual link in an underwater wireless communication network.
This paper presents an intelligent motor speed controller for an earthquake simulator using fuzzy logic algorithm developed inside a programmable logic controller environment. The desired motor speed is obtained using two fuzzy inputs namely, the process error and the rate of process error. These fuzzy inputs are feedback data from the motor drive. Different earthquake intensities were used to test the controller’s performance in real time undergoing different load variations. Experiment results showed that the developed controller is accurate, reliable and robust.
Because of the inherent trade-off between source distortion and channel distortion in video transmission systems, joint optimization between bit-rate and distortion is still a challenging task. In this paper, we propose a method where the bit-rate allocation between source and channel encoder is controlled by the estimated end-to-end distortion at the encoder. The distortion estimation scheme is based on the adaptive forward linear predictor using least-mean square (LMS) algorithm. The forward predictor used the past values of actual end-to-end distortion to estimate the current distortion. The results show good estimate of end-to-end distortion and the proposed scheme improves video quality as compared to a standard rate control of H.264/AVC. The proposed scheme dynamically allocates the source encoder bits based on the estimated distortion.
A fuzzy-based machine vision system was designed to support an industrial sensor based small scale tiger prawn aquaculture system. This system is based on the behavioral movements of tiger prawns as they are subjected to different stress levels such as unsafe values of dissolved oxygen, temperature, salinity and ph. Each parameter is carefully adjusted to trigger a change in the normal behavior or movement of tiger prawns. The change in the behavior of the tiger prawn, as seen by the machine vision system, may serve as a level detection or an alarm system that will aid the sensor based system in terms of monitoring the water quality of the aquaculture environment. This machine vision system may be used to trigger the actuators needed to correct the dangerous water quality parameter back to its safe level. This research is done in real time using two basic web cameras to support industry grade sensors in maintaining the safe water quality level for this small scale habitat.
A new modular relative Jacobian formulation for single end-effector control of combined 3-arm cooperating parallel manipulators is derived. It is based on a previous method of derivation for dual-arm robots, with the same approach of modularity and single end-effector control for combined manipulators. This paper will present this new formulation, as well as investigate task prioritization scheme to verify the claim that a single end-effector controller of combined manipulators will indeed implement a strict task prioritization, by intentionally adding more tasks. In addition, this paper will investigate a claim that the holistic approach to control of combined manipulators affords easier control coordination of each of the stand-alone components. Switching control from an individual manipulator control in the null space to relative control in the tasks space is shown to investigate the smoothness of task execution during switching. Simulation results using Gazebo 2.2.5 running in Ubuntu 14.04 is shown.
This paper presents the development of a computer system for breast cancer awareness and education, particularly, in proper breast self-examination (BSE) performance. It includes the design and development of an artificial intelligent system (AIS) for audio-visual BSE which is capable of computer vision (CV), speech recognition (SR), speech synthesis (SS), and audio-visual (AV) feedback response. The AIS is named BEA, an acronym for Breast Examination Assistant, which acts like a virtual health care assistant that can assist a female user in performing proper BSE. BEA is composed of four interdependent modules: perception, memory, intelligence, and execution. Collectively, these modules are part of an intelligent operating architecture (IOA) that runs the BEA system. The methods of development of the individual subsystems (CV, SR, SS, and AV feedback) together with the intelligent integration of these components are discussed in the methodology section. Finally, the authors presented the results of the tests performed in the system.
Static matrix inverse solving has been studied for many years. In this paper, we aim at solving a dynamic complex-valued matrix inverse. Specifically, based on the artful combination of a conventional gradient neural network and the recently-proposed Zhang neural network, a novel complex-valued neural network model is presented and investigated for computing the dynamic complex-valued matrix inverse in real time. A hardware implementation structure is also offered. Moreover, both theoretical analysis and simulation results substantiate the effectiveness and advantages of the proposed recurrent neural network model for dynamic complex-valued matrix inversion.
This paper proposes an interactive document clustering system, which is designed based on the concept of CMV (coordinated multiple views). An interactive document clustering is used by a user to obtain a set of document groups from a document collection in interactive manner. It is expected to be useful for various tasks such as text mining and document retrieval. As the result of document clustering consists of multiple objects such as clusters (document groups), documents, and words, each of those should be presented to users in different ways. Based on this consideration, the proposed system employs multiple views, each of which is designed for specific object such as document and keyword. A prototype system is implemented on TETDM (Total Environment for Text Data Mining), which is one of environments for developing text data mining tools. As it can provide the mechanism of coordination between modules, we decided to use it for developing the prototype system. The proposed system classifies information to be presented into 4 levels: clusters, document, bag of words, and word, each of which is displayed with different views. Experimental results with test participants show the effectiveness of the proposed system.
The equivalent-input-disturbance (EID) approach is used to deal with the problem of tracking period signals for a plant with input actuator nonlinearities in a repetitive control system. First, an EID estimator is constructed by taking the full advantage of an extended state observer (ESO), the design of the ESO is explained. Next, an EID estimate, which represents the synthetic effect of the nonlinearities, is incorporated into a repetitive control law to compensate for the effect caused by the nonlinearities. This method does not require any prior information about the nonlinearities. It guarantees perfect tracking for periodic reference input and satisfactory compensation for input nonlinearities at the same time. Finally, simulation and experimental results show the effectiveness of the method.
This paper proposes a motion blur length estimation method that is applied to motion blur image restoration. This method applies a cross-correlation algorithm to multi-frame motion-degraded images. In order to find the motion blur parameters, the Radon transform method is used to estimate the motion blur angle. We extract the gray value of pixels around the blur center, calculate the correlation for obtaining motion blur length, and use the Lucy-Richardson iterative algorithm to restore the degraded image. Experiment results show that this method can accurately estimate blur parameters, reduce noise, and obtain better restoration results. The method achieves good results on artificially blurred images and natural images (by the camera shake). The advantage of our algorithm that uses the Lucy-Richardson restoration algorithm compared with the Wiener filtering algorithm is made obvious with less computation time and better restored effects.
Image scrambling is a technique used for confidential storage and transmission as well as for image information hiding. In this study, we propose a multi-channel quantum image scrambling method, which applies both color and geometric transformations of an image. This is a simple and reliable method for transforming a meaningful quantum image into a meaningless or disordered one. We performed two simulation experiments, which demonstrated the efficiency and flexibility of the proposed method. Previous studies mainly developed scrambling strategies for grayscale quantum images, whereas the proposed method is effective for the color image scrambling in the quantum computing domain.
A hybrid modulation strategy (HMS) for a two-stage matrix converter (TSMC) is presented in this paper. According to the variation of voltage transfer ratio, different combinations of modulation modes for rectifier-side converter (RSC) and inverter-side converter (ISC) of TSMC are adopted. Two different current space vector modulation methods are used for RSC to obtain maximum and minor DC voltages. The power loss of TSMC is reduced based on the minor DC voltage. In addition to the linear space vector modulation for ISC, an overmodulation method is presented in order to increase the voltage transfer ratio of TSMC. HMS ensures smooth switching between different modulation modes and makes the best use of the advantage of these modes. Finally, HMS is applied in the case where TSMC is used as an AC-excitation converter for doubly fed induction generator (DFIG) to achieve maximum power point tracking (MPPT). The simulation results confirm the accuracy and feasibility of HMS and the good performance of the MPPT operation of DFIG excited by TSMC.