Reference-Free Machine Vision Inspection of Semiconductor Die Images

Abstract

Advances in electronic technology have made integrated circuits (ICs) the fundamental components in all electronic devices. To increase their production yield by catching defects as early as possible, we need to perform quality assurance on the semiconductor dies during the assembly and packaging processes. A common approach is to employ machine vision to compare a test die with a "known good die". However, difficulties in ensuring identical imaging conditions (such as illumination) are limitations to this die-to-die comparison approach. Instead, in this work we develop a novel reference-free defect detection algorithm for an IC die by analyzing its image. By identifying intrinsic and extrinsic features of various segments in the image, we implement a classification scheme to identify whether the die is defective or not. We rely on the fact that normal ICs contain regular patterns, and the abnormal and irregular regions are classified as potential areas of defects. Experimental results show that the proposed reference-free defect detection algorithm can detect most of the defects from different types of IC dies, and can also correctly classify normal IC dies as non-defective. These results demonstrate the feasibility of the reference-free defect detection approach.

Regularized multiframe phase-shifting algorithm for three-dimensional profilometry

“…Use ϕx; y and ϕ r x; y to recover the surface hx; y from Eqs. (2) and (1). Output: the surface profile hx; y.…”

“…Measuring the surface profile of an object is a very useful procedure in inspection systems as it often sheds light on whether there exists defects or misalignments or even potential connection problems in the forthcoming manufacturing steps. This is particularly challenging when applied to integrated circuit (IC) packages due to two conflicting requirements: a small feature size requiring high precision and the need to a high throughput [1]. There are several general approaches to optical noncontact three-dimensional reconstruction, namely shape from shading (SFS), confocal imaging, stereo, and structured light reconstruction.…”

A three-dimensional imaging system for surface profilometry of moving objects

“…We are most interested in industrial inspection applications, particularly the surface profiling of certain semiconductor components, due to the conflicting requirements to the imaging system and algorithm design: a small feature size that requires a high precision, the need for an inspection on a large field of view (FOV), and a high throughput in the system [3].…”

An illumination-invariant phase-shifting algorithm for three-dimensional profilometry

“…However, surface profiling is particularly challenging when applied to the semiconductor industry, due to two conflicting requirements: a small feature size requiring a high precision, and the need for a high throughput. 1 For certain inspections, explicit three-dimensional reconstruction is not needed and various efficient methods have been proposed. For example, Dong et al proposed a biplanar disparity matrix measure to measure the height of wafer bumps based on a specially designed lighting setup.…”