Automated post bonding inspection by using machine vision techniques

Inspection plays an important role in the semiconductor industry. In this paper, we focus on the inspection task after wire bonding in packaging. The purpose of wire bonding (W/B) is to connect the bond pads with the lead fingers. Two major types of defects are (1) bonding line missing and (2) bonding line breakage. The numbers of bonding lines and bonding balls are used as the features for defect classification. The proposed method consists of image preprocessing, orientation determination, connection detection, bonding line detection, bonding ball detection, and defect classification. The proposed method is simple and fast. The experimental results show that the proposed method can detect the defects effectively.     

3D打印复杂点阵结构的缺陷三维可视化方法

针对3D打印的复杂点阵结构容易出现裂纹、未熔合或孔洞等缺陷,严重影响结构件的功能性能问题,开展了对3D打印的一种复杂点阵结构件的缺陷三维可视化检测方法研究。基于CT图像中结构件内部缺陷的灰度值差异特征,采用集合灰度值法自动识别一类缺陷并分割提取,由光线投射法对分割得到的缺陷序列图像进行三维重构。实验结果表明:所提方法有效获得了点阵结构件内部一种典型缺陷的三维可视图,从三维角度可对缺陷的形状、大小等形貌细节信息进行描述,为进一步分析缺陷对结构性能的影响提供了有力的依据。     

Fast and High-Resolution Optical Inspection System for In-Line Detection and Labeling of Surface Defects

Automated optical inspection systems installed in production lines help ensure high throughput by speeding up inspection of defects that are otherwise difficult to detect using the naked eye. However, depending on the size and surface properties of the products such as micro-cracks on touchscreen panels glass cover, the detection speed and accuracy are limited by the imaging module and lighting technique. Therefore the current inspection methods are still delegated to a few qualified personnel whose limited capacity has been a huge tradeoff for high volume production. In this study, an automated optical technology for in-line surface defect inspection is developed offering high performance in spatial resolution and detection speed for any surface. The inspection system consisting of an LED array which illuminates a wide inspection area on the test object captures scattered light from surface defects using a 12288-pixel line CCD at 12 kHz acquisition rate. A 3.5 \(\mu\) m per pixel resolution of the line CCD provides a detection width capability of at most 43 mm which is equivalent to 147 megapixels image data acquired per second. To handle the large volume of data per acquisition cycle, the data are transmitted from a host CPU to multiple GPU devices where CUDA-based image processing kernels are adopted to perform detection and labeling of surface defects in parallel. The processed data is sent back to the CPU to display user-defined defect maps. 2-D inspection of back-coated flat mirrors, 43 mm x 70 mm\(^{2}\) in size, using a single CCD module and multiple GPU reveals that surface flaws such as bubbles, cracks, and edge defects are detected accurately. The acquisition time to capture and load the data to a CPU is 1.7 s while the processing time to transmit the same data for surface defect detection in a GPU is 248 ms. The latter time scale is considerably faster compared to minute-long computations in solely CPU-based processing algorithm of the same test object. The minimum width of detected surface defects is about 10 \(\mu \)m with true detection rates above 94%. Moreover, the inspection system is easily configurable by tasking multiple CCD imaging modules to different GPU devices to allow inspection of larger test objects. This flexibility can improve both acquisition and detection speeds to boost in-line circuit chips, packaging, and touchscreen panel inspection systems.     

The fatigue threshold, surface condition and fatigue limit of steel wire

To test the hypothesis that fatigue cracks in drawn, pearlitic steel wire propagate from pre-existing surface defects which can be treated as cracks, the fatigue limits of five different wires have been statistically determined. The fatigue thresholds were measured using a new AC potential drop method and the initial defect depths then calculated using the equation a_o=1π(ΔK_th/2σ_e)² and compared with observed values.

For higher strength steel wires (σ_u>1800 MPa) the agreement was very good; for the lower strength steel wire (σ_u=1469 MPa), however, the observed values were smaller than the calculated ones. The reasons for this discrepancy are discussed. A quantitative model is developed from the hypothesis and the influence of residual tensile stresses, decarburization and polishing on the fatigue limit of drawn steel wire is assessed.     

Automated bonding position inspection on multi-layered wire IC using machine vision

No available method can automatically verify the correctness of the wire bonding positions on a multi-layered wire IC. This paper presents a novel method that integrates image processing and wire bonding simulation techniques. The proposed method first takes the IC leadframe image and calculates the lead information before actual wire bonding begins. The wire bonding position information is then generated to simulate the actual wire bonding process. The generated pseudo bonding information is then compared with that from a referential machine. This approach can check the wire bonding position correctness before any actual wire bonding is executed. This approach can fully solve the mal-detection and lost detection problems that may occur in other available methods. The experimental results show that the proposed bonding position check (BPC) method is robust and fast enough for applied multi-layered wire IC inspection synchronously with the wire bonding process.     

雷电流组合波形下碳纤维/环氧树脂基复合材料层压板雷击损伤试验

为了评估碳纤维/环氧树脂基复合材料位于不同飞机雷击分区下的雷击损伤程度及损伤特征,采用A+B、D+B+C和A+B+C+D 3种不同的雷电流波形组合,对两类不同尺寸复合材料层压板试验件（Type1和Type2）进行了模拟雷电流冲击试验,通过目视损伤观察和超声损伤扫描,分析评估了复合材料位于飞机1A区、2A区及1B区时的雷击损伤程度及特征,同时还对含铜网防护碳纤维/环氧树脂基复合材料层压板在不同雷击分区下的雷击防护效果进行了评估。结果表明：不同雷击分区下,碳纤维/环氧树脂基复合材料在雷电流作用下的损伤模式基本一致,包括纤维断裂、基体烧蚀及分层损伤;对于相同类型试验件,1B区对应雷击损伤程度最严重,其次为2A区,1A区损伤程度最小;在相同雷击分区下,长宽比较小的Type1型试验件雷击损伤程度大于长宽比较大的Type2型试验件;0.25 mm厚铜网能够有效对碳纤维/环氧树脂基复合材料进行雷击防护,位于1A区、2A区及1B区的含防护与无防护Type2型试件,前者雷击损伤程度较后者分别下降88.9%、53.9%和68.7%。     

基于 DP 方法的印刷品图像检测技术研究

目的研究印刷品图像质量在线检测及反馈。方法利用机器视觉检测技术、数字图像处理技术,并基于DP方法对标准样张和有缺陷样张进行分割和匹配,根据图像不同的分辨率和所需检测精度来设计图像不同的分割方式,从而进行颜色信息的检测和分析,利用二级识别缺陷分类技术将缺陷进行分类,显示相应的缺陷类型,从而提高印品总体质量。结果与传统的印品质量检测相比,实现了对印刷质量检测的高速度和高精度的要求。结论基于DP方法结合缺陷特征,能够快速、准确地检测出印刷品缺陷。     

基于弱磁的钢丝绳探伤技术研究

依据磁荷分析理论,建立了钢丝绳断丝磁偶极子模型,计算了单丝断口处漏磁场的分布规律。采用正交设计方法对激磁回路参数进行了实验研究,确定了弱磁磁化系统的基本结构与基本参数,实现了磁化与检测的一体化设计。依据弱磁激励情况下钢丝绳缺欠处的漏磁场强度,选择了高灵敏度MR01型磁阻式传感器;给出了多种断丝形态的实测曲线。实验结果表明,基于MR01型磁阻式传感器的钢丝绳磁检测系统,不仅能够实现对小尺寸断口缺陷的有效检测,而且可用于剩磁检测,为钢丝绳探伤技术的发展提供了有价值的科学依据。     

Algorithmic Scheme for Concurrent Detection and Classification of Printed Circuit Board Defects

An ideal printed circuit board (PCB) defect inspection system can detect defects and classify PCB defect types. Existing defect inspection technologies can identify defects but fail to classify all PCB defect types. This research thus proposes an algorithmic scheme that can detect and categorize all 14-known PCB defect types. In the proposed algorithmic scheme, fuzzy c-means clustering is used for image segmentation via image subtraction prior to defect detection. Arithmetic and logic operations, the circle hough transform (CHT), morphological reconstruction (MR), and connected component labeling (CCL) are used in defect classification. The algorithmic scheme achieves 100% defect detection and 99.05% defect classification accuracies. The novelty of this research lies in the concurrent use of CHT, MR, and CCL algorithms to accurately detect and classify all 14-known PCB defect types and determine the defect characteristics such as the location, area, and nature of defects. This information is helpful in electronic parts manufacturing for finding the root causes of PCB defects and appropriately adjusting the manufacturing process. Moreover, the algorithmic scheme can be integrated into machine vision to streamline the manufacturing process, improve the PCB quality, and lower the production cost.     

Improving hydrothermal performance of hybrid nanofluid in double tube heat exchanger using tapered wire coil turbulator

Tapered wire coil insert is proposed as a novel enhancer in the double tube heat exchanger and experimental studies on Al₂O₃ + MgO hybrid nanofluid flowing under the turbulent condition are performed to investigate the hydrothermal characteristics. Effects of using tapered wire coil turbulator and hybrid nanofluid on the hydrothermal behaviors are examined for different coil configurations (Converging (C) type, Diverging (D) type and Conversing-Diverging (C-D) type) and hybrid nanofluid inlet temperatures and volume flow rates. Results show that D-type wire coil insert promotes better hydrothermal performance as compared to C-type and C-D type. Nusselt number and friction factor of hybrid nanofluid using D-type, C-D type and C-type wire coil inserts enhance up to 84%, 71% and 47%, and 68%, 57% and 46%, respectively than that of water in tube without insert. The entropy generation of hybrid nanofluid is lower than that of base fluid in all cases. The thermal performance factor for hybrid nanofluid is found more than one with all inserts. The thermal performance factor is observed a maximum of 1.69 for D-type coil. The study reveals that the hybrid nanofluid and tapered wire coil combination is promising option for improving the hydrothermal characteristics of double pipe heat exchanger.     

One-dimensional-based automatic defect inspection of multiple patterned TFT-LCD panels using Fourier image reconstruction

Thin film transistor-liquid crystal display (TFT-LCD) has been used in a wide range of electronic devices. For large-sized and high-density TFT-LCD panel inspection, a high-resolution line scan is demanded. A TFT-LCD panel image at a fine resolution presents very complicated patterns with less regularity. The paper proposes a non-referential defect detection scheme that directly works on the one-dimensional (1D) line images using the Fourier image reconstruction. The 1D grey-level line image is first divided into small segments, each of the length of the repeated period for a given TFT-LCD panel. The divided segments are then combined as a two-dimensional (2D) image. The frequency components corresponding to the 1D background pattern can be easily identified in the 2D Fourier spectrum. By eliminating the frequency components in the 2D Fourier spectrum that represent the periodic structural pattern of the combined 2D image and then back-transforming the image using the inverse Fourier transform, the 2D Fourier reconstruction process can effectively remove the complicated background pattern and well preserve local anomalies. Experimental results on a number of micro-defects embedded in different patterned regions of TFT-LCD panels show that the proposed method can reliably detect various ill-defined defects.     

The application of synthetic focusing for imaging crack-like defects in pipelines using guided waves

This paper deals with quantifying the performance of a technique for detection, location, and sizing of circumferential crack-like defects in pipelines using synthetically focused guided waves. The system employs a circumferential array of piezoelectric transducer elements. A torsional probing guided wave is excited using the array, which subsequently interacts with the reflecting features of the pipe, such as defects or weld caps. The recorded backscattered signals are synthetically focused to every point of interest in the pipe wall, to form an image of the reflecting features of the pipe. The defect image amplitude is used to estimate the defect depth, and the full width at half maximum of the defect image circumferential profile is used to estimate the circumferential extent of the defect. The imaging system is tested with data from finite element simulations and from laboratory experiments. It is found that reliable sizing of circumferential cracks in finite element simulations and experiments can be achieved if the circumferential extent of the defect is greater than 1.5 lambda_S, where lambda_S is the shear wavelength at the frequency of inspection. This result is theoretically valid for any pipe size, any axial defect location, and any inspection frequency. Amplitude gains of around 18 dB over an unfocused system have been observed experimentally in an 8-inch pipe with a 9 dB SNR improvement.     

Measurement of the size and spacing of standard wire sieves using an image processing system and wavelet transform

A new methodology using image processing and wavelet transform is investigated to measure the size of the wire sieves and their spacing. Wire sieves are used in pharmaceuticals/chemical industries for filtering the grains of chemical powder. Experimental results show measurement that the diameter and spacing of the wire in the sieves can be measured with the accuracy of 1um and uncertainty of 2 μm at 95% confidence level. The method is found suitable for detecting any missing wire or defect like bending or kink in the wire. In this techniques, wavelet transform (Symlet wavelet) analyses the image of sieve in such a way that the discontinuity (cracks, defects, nonuniformity) can be detected more precisely and the spacing/ wire diameter can be measured accurately. This method provides quite can fast and accurate results with ease, in comparison to the existing methods.     

油气输送用埋弧焊管焊接缺陷分析

直缝埋弧焊管焊接完成后,射线检验发现明显连续白色亮带,分析发现其化学成分、力学性能和焊接工艺等方面都满足规范要求,但在金相分析时发现明显的未熔合缺陷。产生未熔合缺陷的原因为四丝焊接时第二丝焊接断弧,送丝轮出现故障,从而停止送丝,最终形成该缺陷。该缺陷处理方法为补焊。     

Effects of bonding frequency on Au wedge wire bondability

This paper studies the effect of bonding frequency on bondability of Au wire on a PCB bond pad. The wire bonding was performed at two different frequencies, 62 kHz and 138 kHz, and at varying bond pad temperatures between 60 and 110 °C. It is shown that the bond strengths of the wires bonded at a high frequency were generally higher than those bonded at a low frequency for all temperatures studied. Two distinct wire failure modes were observed for the wires bonded at the high frequency: the wires with high pull strengths failed at the bond neck, while those with low pull strengths failed mainly within the bonded wire body. This resulted in a large pull strength data scatter, which was explained by the high Q factor of the high frequency transducer. The bondability obtained for the high bond power end was much higher for the high frequency, giving rise to a wider process window in terms of bond power for the high frequency bonding. The wire bonding performance was compared between the “open” and “closed” loop bonding systems. The minimum bond powers required for successful bonds in the closed loop system were significantly lower than those required in the open loop system (e.g. 20–40 mW vs. 110–135 mW at 90 °C). The closed loop system was able to correct the resonance frequency shift, thus maintaining an almost constant bond power during bonding. In the open loop system, in contrast, a high bond power needs to be continuously supplied because of the power drop. This causes an excessive energy to be transmitted to the bonded wire, resulting in weakened wire due to excessive deformation.     

TEM observations of void-lobed defects and the origin of stringers in copper rod and drawn magnet wire

 Utilizing transmission electron microscopy (TEM), we have discovered a defect, which we call a void-lobed defect, in copper precursor rod and in sections of failed copper magnet wire. This defect is the origin of, or basis for, stringers or stringer defects in copper rod and wire, and consists of a contaminated, solidified copper microdroplet which is entrained in the solidifying rod, and is disconnected at its ends in the rod drawing direction, forming voids and elongated void lobes at the ends of the copper inclusion. Utilizing a novel technique to build up fine wires by copper electroplating, we have followed their progress in the magnet-wire drawing process using TEM. Detailed examination of wire failures showed that these defects are forced to the wire axis by successive drawing stages forming coalesced void/debris channels which provide a mechanism for failure. There was no evidence for copper oxides either as stringer components or in connection with debris channels in failed wires. These defects and therefore stringers in precursor rod are created as contaminated or reacted ”copper rain” prior to the solidification stage, and entrained in the rod. The process is usually intermittent and may be reduced or eliminated by properly adjusting the equilibrium chemistry through hydrogen control for oxygen reduction and steam formation; thereby producing high-quality rod for magnet wire produciton. This is an extraordinary example of TEM application in the solution of a contemporary industrial problem which has been otherwise intractable. Received: 2 January 1997 / Accepted: 27 March 1997     

带钢表面缺陷视觉检测系统关键技术研究

为满足钢板表面缺陷在线检测系统宽幅面、高速、高分辨率的检测要求，讨论了基于线阵CCD的钢板表面缺陷视觉检测系统实现的关键技术；优化设计了视觉检测系统的光学照明部分，以检测不同类型的缺陷。通过软件系统的特殊设计，以保证实时在线检测。针对缺陷图像低对比度、高噪声的特点，提出了基于灰度统计特性的图像边缘检测方法，并实现了对缺陷图像的自适应阈值分割。依据图像的缺陷统计特性，定义了缺陷的灰度、几何等特征量，用于缺陷分类。本系统样机已在实验室环境下运行。     

Modelling of the lock-in thermography process through finite element method for estimating the rail squat defects

Squats are a major problem on the world railways. The non-destructive evaluation technique is becoming increasingly attractive in the detection of near surface defects on track. Non-destructive thermal evaluation is one such method of inspection technique that can be used for the detection of near surface defects. Its sub-group of lock-in thermography is under analysis. Lock-in thermography utilizes an infrared camera to detect the thermal waves and then produces a thermal image, which displays the local thermal wave variation in phase or amplitude. There are few studies into the actual experimental representation of complex subsurface defects when concerning lock-in thermography processes. While this may be less of a concern given the purpose of numerical defect characterization to reduce the need for extensive experimental pre-tests, the necessity for (artificial) representations of a defect will inevitably be required for validation. The research outlined in this paper examines the use of 3D finite element modelling (FEM) as a potential flexible tool in simulating the lock-in thermography process for detecting squats in track. In addition, lock-in analysis proved that the correct frequency range had to be selected for the material to detect the defect. As maximum positive and negative phase angles were located at “optimum” frequencies, at certain frequencies lead to minimal phase angle difference to which the defects were not detectable (blind frequency) by using the incorrect testing. The 3D finite element method has advantage for determining the “optimum” thermal excitation frequencies compare with experimental investigation. The experimental results show that 3D FEM models can be used to defect the location and the depth of squats in the railway.     

图像纹理分析在彩色印刷品质量检测中的应用研究

图像处理技术为彩色印刷品的缺陷检测提供了一种新的途径.将图像的纹理分析应用于缺陷检测过程,比较标准图像和待检图像的纹理特征参数就可以分析待检图像细微层次方面的质量,从而检测出印刷缺陷.实验表明,这种缺陷检测方法精度和效率高,满足精细彩色印刷品的要求.