Industrial Anomaly Detection with Skip Autoencoder and Deep Feature Extractor.

Over recent years, with the advances in image recognition technology for deep learning, researchers have devoted continued efforts toward importing anomaly detection technology into the production line of automatic optical detection. Although unsupervised learning helps overcome the high costs associated with labeling, the accuracy of anomaly detection still needs to be improved. Accordingly, this paper proposes a novel deep learning model for anomaly detection to overcome this bottleneck. Leveraging a powerful pre-trained feature extractor and the skip connection, the proposed method achieves better feature extraction and image reconstructing capabilities. Results reveal that the areas under the curve (AUC) for the proposed method are higher than those of previous anomaly detection models for 16 out of 17 categories. This indicates that the proposed method can realize the most appropriate adjustments to the needs of production lines in order to maximize economic benefits.

Applied IrO2 Buffer Layer as a Great Promoter on Ti-Doping V2O5 Electrode to Enhance Electrochromic Device Properties.

Electrochromic devices (ECDs) are a promising material for smart windows that are capable of transmittance variation. However, ECDs are still too expensive to achieve a wide market reach. Reducing fabrication cost remains a challenge. In this study, we inserted an IrO2 buffer layer on Ti-doped V2O5 (Ti:V2O5) as a counter electrode using various Ar/O2 gas flow ratios (1/2, 1/2.5, 1/3 and 1/3.5) in the fabrication process. The buffered-ECD resulted in a larger cyclic voltammetry (CV) area and the best surface average roughness (Ra = 3.91 nm) to promote electrochromic performance. It was fabricated using the low-cost, fast deposition process of vacuum cathodic arc plasma (CAP). This study investigates the influence of the IrO2 buffer/Ti:V2O5 electrode on ECD electrochemical and optical properties, in terms of color efficiency (CE) and cycle durability. The buffered ECD (glass/ITO/WO3/liquid electrolyte/IrO2 buffer/Ti:V2O5/ITO/glass) demonstrated excellent optical transmittance modulation; ∆T = 57% (from Tbleaching (67%) to Tcoloring (10%)) at 633 nm, which was higher than without the buffer (ITO/WO3/liquid electrolyte/Ti:V2O5/ITO) (∆T = 36%). In addition, by means of an IrO2 buffer, the ECD exhibited high coloration efficiency of 96.1 cm2/mC and good durability, which decayed by only 2% after 1000 cycles.

Integrating object detection and image segmentation for detecting the tool wear area on stitched image.

Flank wear is the most common wear that happens in the end milling process. However, the process of detecting the flank wear is cumbersome. To achieve comprehensively automatic detecting the flank wear area of the spiral end milling cutter, this study proposed a novel flank wear detection method of combining the template matching and deep learning techniques to expand the curved surface images into panorama images, which is more available to detect the flank wear areas without choosing a specific position of cutting tool image. You Only Look Once v4 model was employed to automatically detect the range of cutting tips. Then, popular segmentation models, namely, U-Net, Segnet and Autoencoder were used to extract the areas of the tool flank wear. To evaluate the segmenting performance among these models, U-Net model obtained the best maximum dice coefficient score with 0.93. Moreover, the predicting wear areas of the U-Net model is presented in the trend figure, which can determine the times of the tool change depend on the curve of the tool wear. Overall, the experiments have shown that the proposed methods can effectively extract the tool wear regions of the spiral cutting tool. With the developed system, users can obtain detailed information about the cutting tool before being worn severely to change the cutting tools in advance.

High-Performance Complementary Electrochromic Device Based on Iridium Oxide as a Counter Electrode.

In complementary electrochromic devices (ECDs), nickel oxide (NiO) is generally used as a counter electrode material for enhancing the coloration efficiency. However, an NiO film as a counter electrode in ECDs is susceptible to degradation upon prolonged electrochemical cycling, which leads to an insufficient device lifetime. In this study, a type of counter electrode iridium oxide (IrO2) layer was fabricated using vacuum cathodic arc plasma (CAP). We focused on the comparison of IrO2 and NiO deposited on a 5 × 5 cm2 indium tin oxide (ITO) glass substrate with various Ar/O2 gas-flow ratios (1/2, 1/2.5, and 1/3) in series. The optical performance of IrO2-ECD (glass/ITO/WO3/liquid electrolyte/IrO2/ITO/glass) was determined by optical transmittance modulation; ∆T = 50% (from Tbleaching (75%) to Tcoloring (25%)) at 633 nm was higher than that of NiO-ECD (ITO/NiO/liquid electrolyte/WO3/ITO) (∆T = 32%). Apart from this, the ECD device demonstrated a fast coloring time of 4.8 s, a bleaching time of 1.5 s, and good cycling durability, which remained at 50% transmittance modulation even after 1000 cycles. The fast time was associated with the IrO2 electrode and provided higher diffusion coefficients and a filamentary shape as an interface that facilitated the transfer of the Li ions into/out of the interface electrodes and the electrolyte. In our result of IrO2-ECD analyses, the higher optical transmittance modulation was useful for promoting electrochromic application to a cycle durability test as an alternative to NiO-ECD.

A Miniaturized Nickel Oxide Thermistor via Aerosol Jet Technology.

In this study, a miniaturized thermistor sensor was produced using the Aerosol Jet printing process for temperature sensing applications. A nickel oxide nanoparticle ink with a large temperature coefficient of resistance was fabricated. The thermistor was printed with a circular NiO thin film in between the two parallel silver conductive tracks on a cutting tool insert. The printed thermistor, which has an adjustable dimension with a submillimeter scale, operates over a range of 30-250 °C sensitively (B value of ~4310 K) without hysteretic effects. Moreover, the thermistor may be printed on a 3D surface through the Aerosol Jet printing process, which has increased capability for wide temperature-sensing applications.