TSV三维封装缺陷激光主动检测分类与量化

硅通孔(TSV)三维封装因其独特工艺而受到广泛关注,然而其内部缺陷无法被直接观测成为目前的检测难题。为了对TSV三维封装内部缺陷进行检测,提出了一种基于激光主动激励的内部缺陷分类与量化方法。通过红外激光主动热源施加到TSV三维封装结构表面,激发TSV内部缺陷的外部温度分布响应,通过理论与仿真分析,掌握缺陷特征在主动激励下的表现规律；构建卷积神经网络对缺陷样本信息进行训练,实现内部缺陷的分类识别与量化。试验表明,该方法能在不损坏样品的前提下有效对内部缺陷进行识别分类及量化,准确率可达95.56%。

Active Laser Detection Classification and Quantification of Defects in TSV 3D Packaging

Through-silicon via three-dimensional packaging has received widespread attention due to its unique process. However, its internal defects cannot be directly observed, which has become a current detection problem. In order to detect internal defects in TSV three-dimensional packaging, an internal defect classification and quantification method based on laser active excitation is proposed. The infrared laser active heat source is applied to the surface of the TSV three-dimensional packaging structure to stimulate the external temperature distribution response of the internal defects of the TSV. Through theoretical and simulation analysis, the behavior of the defect characteristics under active excitation is grasped; a convolutional neural network is then established to train the defect sample information, achieving the classification and quantification of internal defects. The experiments demonstrate that this method can effectively identify, classify, and quantify internal defects without damaging the samples, with an accuracy rate of up to 95.56%.