A Kinect-Based Segmentation of Touching-Pigs for Real-Time Monitoring.

Segmenting touching-pigs in real-time is an important issue for surveillance cameras intended for the 24-h tracking of individual pigs. However, methods to do so have not yet been reported. We particularly focus on the segmentation of touching-pigs in a crowded pig room with low-contrast images obtained using a Kinect depth sensor. We reduce the execution time by combining object detection techniques based on a convolutional neural network (CNN) with image processing techniques instead of applying time-consuming operations, such as optimization-based segmentation. We first apply the fastest CNN-based object detection technique (i.e., You Only Look Once, YOLO) to solve the separation problem for touching-pigs. If the quality of the YOLO output is not satisfied, then we try to find the possible boundary line between the touching-pigs by analyzing the shape. Our experimental results show that this method is effective to separate touching-pigs in terms of both accuracy (i.e., 91.96%) and execution time (i.e., real-time execution), even with low-contrast images obtained using a Kinect depth sensor.

Replacement Condition Detection of Railway Point Machines Using an Electric Current Sensor.

Detecting replacement conditions of railway point machines is important to simultaneously satisfy the budget-limit and train-safety requirements. In this study, we consider classification of the subtle differences in the aging effect-using electric current shape analysis-for the purpose of replacement condition detection of railway point machines. After analyzing the shapes of after-replacement data and then labeling the shapes of each before-replacement data, we can derive the criteria that can handle the subtle differences between "does-not-need-to-be-replaced" and "needs-to-be-replaced" shapes. On the basis of the experimental results with in-field replacement data, we confirmed that the proposed method could detect the replacement conditions with acceptable accuracy, as well as provide visual interpretability of the criteria used for the time-series classification.

Depth-Based Detection of Standing-Pigs in Moving Noise Environments.

In a surveillance camera environment, the detection of standing-pigs in real-time is an important issue towards the final goal of 24-h tracking of individual pigs. In this study, we focus on depth-based detection of standing-pigs with "moving noises", which appear every night in a commercial pig farm, but have not been reported yet. We first apply a spatiotemporal interpolation technique to remove the moving noises occurring in the depth images. Then, we detect the standing-pigs by utilizing the undefined depth values around them. Our experimental results show that this method is effective for detecting standing-pigs at night, in terms of both cost-effectiveness (using a low-cost Kinect depth sensor) and accuracy (i.e., 94.47%), even with severe moving noises occluding up to half of an input depth image. Furthermore, without any time-consuming technique, the proposed method can be executed in real-time.

A fully roll-to-roll gravure-printed carbon nanotube-based active matrix for multi-touch sensors.

Roll-to-roll (R2R) printing has been pursued as a commercially viable high-throughput technology to manufacture flexible, disposable, and inexpensive printed electronic devices. However, in recent years, pessimism has prevailed because of the barriers faced when attempting to fabricate and integrate thin film transistors (TFTs) using an R2R printing method. In this paper, we report 20 × 20 active matrices (AMs) based on single-walled carbon nanotubes (SWCNTs) with a resolution of 9.3 points per inch (ppi) resolution, obtained using a fully R2R gravure printing process. By using SWCNTs as the semiconducting layer and poly(ethylene terephthalate) (PET) as the substrate, we have obtained a device yield above 98%, and extracted the key scalability factors required for a feasible R2R gravure manufacturing process. Multi-touch sensor arrays were achieved by laminating a pressure sensitive rubber onto the SWCNT-TFT AM. This R2R gravure printing system overcomes the barriers associated with the registration accuracy of printing each layer and the variation of the threshold voltage (Vth). By overcoming these barriers, the R2R gravure printing method can be viable as an advanced manufacturing technology, thus enabling the high-throughput production of flexible, disposable, and human-interactive cutting-edge electronic devices based on SWCNT-TFT AMs.

Scalability of carbon-nanotube-based thin film transistors for flexible electronic devices manufactured using an all roll-to-roll gravure printing system.

To demonstrate that roll-to-roll (R2R) gravure printing is a suitable advanced manufacturing method for flexible thin film transistor (TFT)-based electronic circuits, three different nanomaterial-based inks (silver nanoparticles, BaTiO3 nanoparticles and single-walled carbon nanotubes (SWNTs)) were selected and optimized to enable the realization of fully printed SWNT-based TFTs (SWNT-TFTs) on 150-m-long rolls of 0.25-m-wide poly(ethylene terephthalate) (PET). SWNT-TFTs with 5 different channel lengths, namely, 30, 80, 130, 180, and 230 µm, were fabricated using a printing speed of 8 m/min. These SWNT-TFTs were characterized, and the obtained electrical parameters were related to major mechanical factors such as web tension, registration accuracy, impression roll pressure and printing speed to determine whether these mechanical factors were the sources of the observed device-to-device variations. By utilizing the electrical parameters from the SWNT-TFTs, a Monte Carlo simulation for a 1-bit adder circuit, as a reference, was conducted to demonstrate that functional circuits with reasonable complexity can indeed be manufactured using R2R gravure printing. The simulation results suggest that circuits with complexity, similar to the full adder circuit, can be printed with a 76% circuit yield if threshold voltage (Vth) variations of less than 30% can be maintained.