Deep learning Optical Character Recognition in PCB Dark Silk Recognition

For Automatic Optical Inspection (AOI) machines that were introduced to Printed Circuit Board market more than five years ago, illumination technique and light devices are outdated. Images captured by old AOI machines are not easy to be recognized by typical optical character recognition (OCR) algorithms, especially for dark silk. How to effectively increase silk recognition accuracy is indispensable for improving overall production efficiency in SMT plant. This paper uses fine tuned Character Region Awareness for Text Detection (CRAFT) method to build model for dark silk recognition. CRAFT model consists of a structure similar to U-net, followed by VGG based convolutional neural network. Continuous two-dimensional Gaussian distribution was used for the annotation of image segmentation. CRAFT model is good at recognizing different types of printed characters with high accuracy and transferability. Results show that with the help of CRAFT model, accuracy for OK board is 95% (error rate is 5%), and accuracy for NG board is 100% (omission rate is 0%).

Convolutional Neural Network and Bayesian Gaussian Process in Driving Anger Recognition

With the development of motorization, road traffic crashes have become the leading cause of death in many countries. Among roadway traffic crashes, almost 90% of accidents are related to driver behaviors, wherein driving anger is one of the most leading causes to vehicle crash-related conditions. To some extent, angry driving is considered more dangerous than typical driving distraction due to emotion agitation. Aggressive driving behaviors create many kinds of roadway traffic safety hazards. Mitigating potential risk caused by road rage is essential to increase the overall level of traffic safety. This paper puts forward an integrated computer vision model composed of convolutional neural network in feature extraction and Bayesian Gaussian process in classification to recognize driver anger and distinguish angry driving from natural driving status. Histogram of gradients (HOG) was applied to extract facial features. Convolutional neural network extracted features on eye, eyebrow, and mouth, which are considered most related to anger emotion. Extracted features with its probability were sent to Bayesian Gaussian process classier as input. Integral analysis on three extracted features was conducted by Gaussian process classifier and output returned the likelihood of being anger from the overall study of all extracted features. An overall accuracy rate of 86.2% was achieved in this study. Tongji University 8-Degree-of-Freedom driving simulator was used to collect data from 30 recruited drivers and build test scenario.

Analysis of the Relationship between the Number of Traffic Accidents and the Traffic Flow &Section Location in Extra Long Tunnel

In order to study the influence of the traffic characteristics on traffic accidents in extra long tunnel, the main measurement indicators of traffic flow during the time of traffic accidents are matched with the accident information to form a data set of the number of traffic accidents and the hourly traffic flow of the accident. Vehicle ratio and the number of accidents are mainly used as the characteristic indicators of traffic flow. At the same time, the longitudinal distribution law of the average speed of traffic flow and the number of traffic accidents in the extra long tunnel is studied. Based on the superposition principle, the extra long tunnel is divided into 5 traffic safety zones. This paper analyzes the distribution of time, morphology, cause of accident, and other characteristics in different traffic safety zones, finding that the shape of traffic accidents in extra long tunnel is mainly rear-end collisions. Improper operation and illegal lane changes are the main causes of accidents.

Mechanical Analysis and Performance Optimization for the Lunar Rover’s Vane-Telescopic Walking Wheel

It is well-known that optimizing the wheel system of lunar rovers is essential.However,this is a difficult task due to the complex terrain of the moon and limited resources onboard lunar rovers.In this study,an experimental prototype was set up to analyze the existing mechanical design of a lunar rover and improve its performance.First,a new vane-telescopic walking wheel was proposed for the lunar rover with a positive and negative quadrangle suspension,considering the complex terrain of the moon.Next,the performance was optimized under the limitations of preserving the slope passage and minimizing power consumption.This was achieved via analysis of the wheel force during movement.Finally,the effectiveness of the proposed method was demonstrated by several simulation experiments.The newly designed wheel can protrude on demand and reduce energy consumption;it can be used as a reference for lunar rover development engineering in China.

Fully Connected Convolutional Neural Network in PCB Soldering Point Inspection

In Electronics Manufacturing Services (EMS) industry, Printed Circuit Board (PCB) inspection is tricky and hard, especially for soldering point inspection due to the extremely tiny size and inconsistent appearance for uneven heating in reflow soldering process. Conventional computer vision technique based on OpenCV or Halcon usually cause false positive call for originally good soldering point on PCB because OpenCV or Halcon use the pre-defined threshold in color proportion for deciding whether the specific soldering point is OK or NG (not good). However, soldering point forms are various after heating in reflow soldering process. This paper puts forward a VGG structure deep convolutional neural network, which is named SolderNet for processing soldering point after reflow heating process to effectively inspect soldering point status, reduce omission rate and error rate, and increase first pass rate. SolderNet consists of 11 hidden convolution layers and 3 densely connected layers. Accuracy reports are divided into OK point recognition and NG point recognition. For OK soldering point recognition, 92% is achieved. For NG soldering point recognition, 99% is achieved. The dataset is collected from KAGA Co. Ltd Plant in Suzhou. First pass rate at KAGA plant is increased from 25% to 80% in general.

High thermoelectric figure of merit ZT>1 in SnS polycrystals

Eco-friendly tin sulfide(SnS)has attracted increasing attention in the thermoelectric community because of its elemental abundance and analogous crystal structure to SnSe as a new thermoelectric material.However,so far no high dimensionless thermoelectric figure of merit ZT>1 was reported in SnS polycrystals.This work found an effective strategy for enhancing the thermoelectric performance of ptype polycrystalline SnS by Ag doping and vacancy engineering,leading to three orders of magnitude increase in carrier concentration and optimized effective mass and carrier mobility.As a result of the enhanced electrical conductivity,three times higher power factor ~3.85 μW/cm K^(2) at 877 K is realized in Sn_(0.995)Ag_(0.005)S sample.Interestingly,nanostructuring with Ag nano-precipitates were formed in the Agdoped SnS sample.Moreover,with introducing Sn vacancies in the crystal structure of Sn_(0.995-vac)Ag_(0.005)S,the power factor further enhanced to~4.25 μW/cm K^(2).In addition to the low-frequency phonons scattering by Ag nano-precipitates,dislocations strengthens the scattering of mid-frequency phonon,leading to an ultralow lattice thermal conductivity <0.5 W/m K above 800 K and a record high ZT up to 1.1 at 877 K in Sn_(0.99)Ag_(0.005)S polycrystals.

Enhanced thermoelectric performance of Na-doped Pb Te synthesized under high pressure

Despite an effective p-type dopant for PbTe, the low solubility of Na limits the fully optimization of thermoelectric properties of Na-doped PbTe. In this work, Na-doped PbTe was synthesized under high pressure. The formation of the desired rocksalt phase with substantially increased Na content leads to a high carrier concentration of 3.2×10^20 cm^-3 for Na0.03Pb0.97Te. Moreover, dense in-grain dislocations are identified from the microstructure analysis. Benefited from the improved power factor and greatly suppressed lattice thermal conductivity, the maximal ZT of 1.7 is achieved in the optimal Na0.03Pb0.97Te. Current work thus designates the advantage of high pressure in synthesizing PbTe-based thermoelectric materials.

豆蛋白与丝素复合纳米纤维膜的喷纺制备、结构与性能

对丰富的绿色天然资源的合理开发并应用于生物医药与组织工程领域的研究是当今的一个科学热点,具有重要的现实意义.本文利用喷纺技术成功制备了具有生物相容性与可生物降解的豆蛋白与丝素复合纳米纤维膜.通过扫描电子显微观察、拉曼光谱和傅里叶变换红外光谱以及X射线衍射分析,表征了材料的表观形貌和结构变化与相互作用;利用差示扫描量热分析和热重分析,结合高分子共混模型,考察了两相材料的喷纺共混性,评价了复合膜的保水能力、体外降解与细胞相容性.结果表明,丝素和豆蛋白分子链之间主要存在氢键相互作用,具有热力学相容性,并呈现纳米纤维形貌、低孔隙率和优良的热性质.随着豆蛋白含量的增加,复合纳米纤维膜中的β折叠含量逐渐增加,玻璃化转变温度降低,比热容增量略有增大,保水能力增强,酶降解速度加快.同时,与纯丝素纤维膜相比,豆蛋白/丝素复合纤维膜能更好地为细胞附着和增殖提供有利环境,可以用作敷料与载药系统,特别是含60%豆蛋白的复合纤维膜显示出最强的细胞活性.本研究可以为促进丝素和豆蛋白等天然材料的纳米制备提供新方法,并为绿色天然资源的广泛应用提供实验和理论参考.

A sound velocity method for determining isobaric specific heat capacity

Isobaric specific heat capacity(Cp)is an important parameter not only in physics but also for most materials.Its accurate measurement is particularly critical for performance evaluation of thermoelectric materials,but the experiments by differential scanning calorimetry(DSC)often lead to large uncertainties in the measurements,especially at elevated temperatures.In this study,we propose a simple method to determine Cp by measuring the sound velocity(υ)based on lattice vibration and expansion theory.The relative standard error of theυis smaller than 1%,showing good accuracy and repeatability.The calculated Cp at elevated temperature(>300 K)increases slightly with increasing temperature due to the lattice expansion,which is more reasonable than the Dulong–Petit value.

(Bi,Sb)_(2)Te_(3)/SiC nanocomposites with enhanced thermoelectric performance:Effect of SiC nanoparticle size and compositional modulation

Fabrication of nanoparticle-dispersed composites is an effective strategy for enhancing the performance of thermoelectric materials,and in particular SiC nanoparticles have been often used to create composites with Bi_(2)Te_(3)-based applied thermoelectric materials.However,the effect of particle size on the thermoelectric performance is unclear.This work systematically investigated the electrical and thermal properties of a series of(Bi,Sb)_(2)Te_(3)-based nanocomposites containing dispersed SiC nanoparticles of different sizes.It was found that particle size has a significant impact on the electrical properties with smaller SiC nanoparticles giving rise to higher electrical conductivity.Even though the dispersed SiC nanoparticles enhanced the Seebeck coefficient,no apparent dependence of the enhancement on the particle size was observed.It was also found that smaller SiC nanoparticles scatter phonons to some extent while the larger nanoparticles contribute to increased thermal conductivity.Eventually,the highest ZT value of 1.12 was obtained in 30 nm-SiC dispersed sample,corresponding to an increase by 18%from 0.95 for the matrix made from commercial scraps,and then the ZT was further boosted to 1.33 by optimizing the matrix composition and expelling excess Te during the optimized spark plasma sintering process.This work proves that the dispersion of smaller SiC nanoparticles in p-type(Bi,Sb)_(2)Te_(3) materials is more effective than the dispersion of larger nanoparticles.In addition,it is revealed that additional compositional and/or processing optimization is vital and effective for obtaining further performance enhancement for nanocomposites of SiC nanoparticles dispersed in(Bi,Sb)_(2)Te_(3).

Enhanced thermoelectric performance in MnTe due to doping and insitu nanocompositing effects by Ag2S addition

Extremely low lattice thermal conductivity is always the pursuit of thermoelectric materials research.In this work,we reported an exceptional effect of Ag2S addition in MnTe,an emerging promising midtemperature thermoelectric material,to enable the realization of minimum lattice thermal conductivity,namely-0.4 Wm^(-1) K^(-1).Such a low lattice thermal conductivity is guaranteed by the incorporation of in-situ formed Ag rich phase(Ag2Te)with ultralow lattice thermal conductivity and further scattering of phonons from the partial doping effects induced point defects and boundaries between various phases.Apart from the dramatically decreased lattice thermal conductivity,the partial doping of Ag and S simultaneously enhance the electrical conductivity,further contributing to enhanced thermoelectric performance.Meanwhile,an inverse sign of Seebeck and Hall coefficient was observed and rationalized by the influence of highly electron-conductive Ag_(2)Te phase.Thanks to the synergetic modulation of electrical and thermal transport properties by in-situ formed composite,a high ZT value of 1.1 was achieved in MnTe based thermoelectric materials,which also demonstrates the importance of compositing approaches to design state-of-the-art thermoelectric materials.