Nonlinear color blending, which is advanced blending indicated by blend modes such as "overlay" and "multiply," is extensively employed by digital creators to produce attractive visual effects. To enjoy such flexible editing modalities on existing bitmap images like photographs, however, creators need a fast nonlinear blending algorithm that decomposes an image into a set of semi-transparent layers. To address this issue, we propose a neural-network-based method for nonlinear decomposition of an input image into linear and nonlinear alpha layers that can be separately modified for editing purposes, based on the specified color palettes and blend modes. Experiments show that our proposed method achieves an inference speed 370 times faster than the state-of-the-art method of nonlinear image unblending, which uses computationally intensive iterative optimization. Furthermore, our reconstruction quality is higher or comparable than other methods, including linear blending models. In addition, we provide examples that apply our method to image editing with nonlinear blend modes.

This paper presents a FPGA-based analog-digital hybrid real-time simulator (HRTS)with a new state space representation method. In the proposed method, electro-magnetic transient models with switches are represented by parallel ordinary differential equation (ODE)based on a block-diagonal decomposition of matrix. This paper also proposes a new update rule of state space form in order to accurately calculate with limited resources. The proposed method can be implemented with low costs of computation, and enables accurate computation with a less size fixed point of Field-Programmable Gate Array (FPGA). We implemented a doubly fed induction generator (DFIG)based wind turbine model on the FPGA using the proposed method and connected it to the analog simulator via a current source. By comparing with MATLAB/Simulink references, the effectiveness of the proposed method is illustrated.

Motivation: The aim of precision medicine is to harness new knowledge and technology to optimize the timing and targeting of interventions for maximal therapeutic benefit. This study explores the possibility of building AI models without precise pixel-level annotation in prediction of the tumor size, extrathyroidal extension, lymph node metastasis, cancer stage and BRAF mutation in thyroid cancer diagnosis, providing the patients' background information, histopathological and immunohistochemical tissue images. Results: A novel framework for objective evaluation of automatic patient diagnosis algorithms has been established under the auspices of the IEEE International Symposium on Biomedical Imaging 2017- A Grand Challenge for Tissue Microarray Analysis in Thyroid Cancer Diagnosis. Here, we present the datasets, methods and results of the challenge and lay down the principles for future uses of this benchmark. The main contributions of the challenge include the creation of the data repository of tissue microarrays; the creation of the clinical diagnosis classification data repository of thyroid cancer; and the definition of objective quantitative evaluation for comparison and ranking of the algorithms. With this benchmark, three automatic methods for predictions of the five clinical outcomes have been compared, and detailed quantitative evaluation results are presented in this paper. Based on the quantitative evaluation results, we believe automatic patient diagnosis is still a challenging and unsolved problem. Availability and implementation: The datasets and the evaluation software will be made available to the research community, further encouraging future developments in this field. (http://www-o.ntust.edu.tw/cvmi/ISBI2017/). Contact: [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.

We present a video annotation system called ``AnnoTone', which can embed various contextual information describing a scene, such as geographical location. Then the system allows the user to edit the video using this contextual information, enabling one to, for example, overlay with map or graphical annotations. AnnoTone converts annotation data into high-frequency audio signals (which are inaudible to the human ear), and then transmits them from a smartphone speaker placed near a video camera. This scheme makes it possible to add annotations using standard video cameras with no requirements for specific equipment other than a smartphone. We designed the audio watermarking protocol using dual-tone multi-frequency signaling, and developed a general-purpose annotation framework including an annotation generator and extractor. We conducted a series of performance tests to understand the reliability and the quality of the watermarking method. We then created several examples of video-editing applications using annotations to demonstrate the usefulness of Annotone, including an After Effects plug-in.

This paper introduces an input and output device that enables illumination, bi-directional data communication, and position sensing on a soft cloth. This "LightCloth" is woven from diffusive optical fibers. Since the fibers are arranged in parallel, the cloth has one-dimensional position information. Sensor-emitter pairs attached to bundles of contiguous fibers enable bundle-specific light input and output. We developed a prototype system that allows full-color illumination and 8-bit data input by infrared signals. We present as an application a chair with a LightCloth cover whose illumination pattern is specified using an infrared light pen. Here we describe the implementation details of the device and discuss possible interactions using the device.

This paper introduces optimal operational planning of energy plants via the ε constrained differential evolution. In order to generate optimal operational planning for energy plants, startup/shutdown status and/or input/output values of the facilities at each control interval should be determined. The problem can be formulated as a large-scale mixed-integer nonlinear problem (MINLP). Metaheuristics (MHs) is one of the solutions for MINLP. If the formulated MINLP has various equality and inequality constraints, it remains difficult to solve it without parameters tuning. In this paper, to overcome the difficulty, we propose an improved differential evolution approach using the ε constrained method for MINLP. Results show the effectiveness of the proposed method compared with conventional methods.

Sensor networks using mobile robots have recently been proposed to provide more flexible and efficient sensing. However, mobile robots in a practical world have many accidents with actuator or communication devices especially in their traveling to the destination. Moreover, in the practical world, there are many obstacles and bad conditioned area for both wireless and movement of mobile robots. In this paper, we introduce a practical scheme for delivering data with concerning a practical world issues.

Sensor networks using mobile robots have been proposed to provide more flexible and efficient sensing. One of the key challenges under this paradigm is a management of transferring a huge amount of data. In this paper, we propose a model and protocol for session control exploiting controlled mobility in multi-robot sensor networks (MRSNs) to transfer the huge amount of data efficiently. Since a communication in MRSNs is mainly provided by a physical movement of the mobile node, the performance of the session heavily depends on this movement. However, if a node initiating the session can utilize sufficient amount of nodes, the performance of the session significantly depends on the wireless communication. In this paper, we propose a suitable method of session control which is determined the balance between movement and wireless communications based on local information such as the location, number, maximum-velocity and buffer-size of surrounding nodes. WISER/s also takes a system policy of either delay minimization or energy efficiency into consideration. In this paper, we describe a method of how to initiate, terminate and manage the session including dynamic selection of optimal nodes' formation. We evaluate the performance of WISER/s using mobile robots which are equipped with MICA2 mote and comparing with non-optimized method. The experimental results demonstrate that WISER/s achieves better performance than non-optimized method.

This paper presents an automatic tuning method of model predictive control (MPC) using particle swarm optimization (PSO). Although conventional PID is difficult to treat constraints and future plant dynamics, MPC can treat this issues and practical control can be realized in various industrial problems. One of the challenges in MPC is how control parameters can be tuned for various target plants and usage of PSO for automatic tuning is one of the solutions. The numerical results show the effectiveness of the proposed PSO-based automatic tuning method

In this paper we propose a design of application programming interfaces (APIs) for configuration of multi-robot sensor networks (MRSNs). Such networks have unique aspects: intentional mobility and delay tolerance. These aspects limit the applicability of conventional approaches proposed for mobile ad-hoc networks (MANETs). APIs for configuration of MRSNs must take these aspects into account. In addition, an application program should be transparent from changing of physical devices such as wireless devices. In this paper we describe the design and prototype implementation of APIs for configuration of MRSNs.

Wireless sensor networks have traditionally focused on low duty-cycle applications where sensor data are reported periodically in the order of seconds or even longer. This is due to typically slow changes in physical variables, the need to keep node costs low and the goal of extending battery lifetime. However, there is a growing need to support real-time streaming of audio and/or low-rate video even in wireless sensor networks for use in emergency situations and short-term intruder detection. In this paper, we present FireFly, a time-synchronized sensor network platform for real-time data streaming across multiple hops. FireFly is composed of several integrated layers including specialized low-cost hardware, a sensor network operating system, a real-time link layer and network scheduling which together provide efficient support for applications with timing constraints. In order to achieve high end-to-end throughput, bounded latency and predictable lifetime, we employ hardware-based time synchronization. Multiple tasks including audio sampling, networking and sensor reading are scheduled using the nano-RK RTOS. We have implemented RT-Link, a TDMA-based link layer protocol for message exchange on well-defined time slots and pipelining along multiple hops. We use this platform to support 2-way audio streaming concurrently with sensing tasks. For interactive voice, we investigate TDMA-based slot scheduling with balanced bi-directional latency while meeting audio timeliness requirements. Finally, we describe our experimental deployment of 42 nodes in a coal mine, and present measurements of the end-to-end throughput, jitter, packet loss and voice quality