Persistent feature detection

Abstract

Methods are presented for improved detection of persistent or systematic defects induced during the manufacture of a product. In particular, the methods are directed to the detection of defects induced systematically in the manufacture of photovoltaic cells and modules. Images acquired from a number of samples are combined, enhancing the systematic defects and suppressing random features such as variations in material quality. Once a systematic defect is identified, steps can be taken to locate and rectify its cause.

1 . An apparatus for detecting systematic features in products being manufactured on a process line, said apparatus comprising:
 a camera configured to acquire images of a plurality of products, each image being of substantially the same area on each of said products; and   a processor configured to combine said images and obtain a super-image, process said super-image to identify regions with strong signals, and identify said regions as being indicative of systematic features in said products.   
     
     
         2 . An apparatus for detecting systematic features in products being manufactured on a process line, said apparatus comprising:
 a camera configured to acquire images of a plurality of products, each image being of substantially the same area on each of said products; and   a processor configured to combine said images and obtain a super-image, process said super-image to identify regions with strong signals, compare said regions with known features in said area, and identify those of said regions that do not correspond to said known features as being indicative of systematic features in said products.   
     
     
         3 . An apparatus according to  claim 1 , wherein the processor is configured to correct said images to remove known features in said area. 
     
     
         4 . An apparatus according to  claim 1 , wherein the processor is configured to combine said images by summing, averaging or obtaining a median. 
     
     
         5 . An apparatus according to  claim 1 , wherein the processor is configured to identify said regions with strong signals by comparison with a predetermined threshold. 
     
     
         6 . An apparatus according to  claim 1 , further comprising an optical source of wavelength and intensity suitable for generating photoluminescence from said products, such that said camera acquires photoluminescence images. 
     
     
         7 . An apparatus according to  claim 1 , further comprising a source of electrical power for electrically exciting said products to emit electroluminescence, such that said camera acquires electroluminescence images. 
     
     
         8 . An apparatus according to  claim 1 , further comprising a source of visible or infrared radiation, such that said camera acquires reflection or transmission images in the visible or infrared spectral regions. 
     
     
         9 . An apparatus according to  claim 1 , wherein said apparatus is provided at multiple stages in a process line manufacturing said products, to identify stages in said process line where said systematic features are introduced. 
     
     
         10 . An apparatus according to  claim 1 , wherein said apparatus is configured to detect systematic features in photovoltaic wafers, cells or modules. 
     
     
         11 . An apparatus according to  claim 10 , wherein said systematic features are cracks in said photovoltaic wafers, cells or modules. 
     
     
         12 . An apparatus according to  claim 10 , wherein the processor is configured to process said super-image to determine a background signal level in portions away from said regions with strong signals, and use said background signal level to predict the performance of said photovoltaic wafers, cells or modules. 
     
     
         13 . An apparatus according to  claim 1 , wherein the processor is configured to combine said images in the Fourier, Discrete Cosine Transform, Hadamard or Wavelet domain. 
     
     
         14 . An apparatus according to  claim 1 , wherein the processor is configured to monitor statistical variances from expected values for regions of said images. 
     
     
         15 . An apparatus according to  claim 1 , wherein the processor is configured to align said images prior to combining said images. 
     
     
         16 . An apparatus according to  claim 1 , wherein the processor is configured to identify the type or nature of said systematic features. 
     
     
         17 . An apparatus for detecting process-induced defects in products being manufactured on a process line, said apparatus comprising:
 a camera configured to acquire images of corresponding areas of a plurality of products; and   a processor configured to cumulate said images and produce a cumulative image and identify signals in said cumulative image which exceed a predetermined threshold, thereby determining said process-induced defects.   
     
     
         18 . An apparatus for detecting process-induced defects in products being manufactured on a process line, said apparatus comprising:
 a camera configured to acquire images of corresponding areas of a plurality of products; and   a processor configured to cumulate said images and produce a cumulative image, identify signals in said cumulative image which exceed a predetermined threshold, thereby determining one or more defect candidates, and exclude from said candidates any known features, thereby determining said process-induced defects.   
     
     
         19 . An apparatus according to  claim 17 , wherein the processor is configured to correct said images by removing signals in each said image which correspond to any known features. 
     
     
         20 . In a process line for manufacturing products, an apparatus for detecting systematic features in products being manufactured on said process line, said apparatus comprising:
 a camera configured to acquire images of a plurality of said products, each image being of substantially the same area on each of said products; and   a processor configured to combine said images and obtain a super-image, process said super-image to identify regions with strong signals, and identify said regions as being indicative of systematic features in said products.   
     
     
         21 . In a process line for manufacturing products, an apparatus for detecting process-induced defects in products being manufactured on said process line, said apparatus comprising:
 a camera configured to acquire images of corresponding areas of a plurality of said products; and   
       a processor configured to cumulate said images and produce a cumulative image, and identify signals in said cumulative image which exceed a predetermined threshold, thereby determining said process-induced defects.