Image Sensor Lab - Image Sensor Evaluation for CMOS Image Sensors

Home	Contact Us	Store	Cart

Instruments

Image Sensor Lab

Interface Spec

ISL for R&D

ISL Store

Product Sheet (pdf)

White Paper (pdf)

Stepper Motor Control

USB to I2c and SPI

Tech Support

Instrument Store

Contact Sales

Image Sensor Lab™

Test System for CMOS Digital Image Sensors


		ISL for Research and Development
			Focus Accuracy
			Photo Response Uniformity
			Light Field Dead and Defective Pixel Detection
			Dark Field Dead and Defective Pixel Detection
			Luminance Uniformity
			Dust and Particle Detection
			Macbeth Color Accuracy
			Lens Shading Correction
			Dim Region Detection


		Print This Page
		Email This Page


	Image Sensor Lab
	From: $4695

or contact sales
to order

Bayer Light Field Dead and Defective Pixel Detection

Overview

Light Field dead and defective pixel testing is performed by first exposing the image sensor field of view to a uniform light field. Two types or classes of outlier are then identified

Dead Pixels - Any pixel with intensity below a specified % of mean is deemed a dead pixel.

Defective Pixels - Any pixel that deviates from the mean light field intensity by more than a specified % of mean is deemed defective.

Image Sensor Lab provides slider controls and real-time dead and defect detection table output. An intensity histogram graph with the dead and defective pixel limits overlaid is also provided for improved user feedback.

The defective pixel test is more stringent than the dead pixel test. Many CMOS image sensors have built in compensation for dead and defective pixels but are limited to the total number of bad pixels that can be corrected. Individual, isolated dead or defective pixels are usually tolerated, but since multiple defective pixels in a small region would be noticeable or may not be correctable by the built-in processing, limits are typically set not only on the total number of dead or defective pixels but also on the number of defective rows, columns, and clusters. Image Sensor Lab provides defective row, column and cluster detection and reporting and allows the user to specify the detection criteria as shown at left.

Light Field Dead and Defective Pixel Detection Algorithm

The Bayer Light Field Dead and Defective Pixel detection algorithm is a simple thresholding operation that classifies any pixel with intensity below a user-specified limit as a dead pixel. The test also detects defective pixels, which are defined are pixels that exceed a user-defined percentage threshold above or below the local mean.

The algorithm also allows user configurable maximum number of defect speci-fications. The maximum number of allowable dead or defective pixels of one color in a 5 by 5 regions or any color in a 3 by 2 region can be specified as can the maximum number of defective rows or columns.

A background subtraction image processing step can be selected that can help eliminate lens shading effects, when using this test as a measure of the pixel noise in the image. The correction algorithm uses an Nth order filter and has user configurable kernel size and percentile threshold algorithm inputs.

Light Field Dead and Defective Pixel Input Parameters

*Name*	*Description*

Input Image	A Bayer image type is required for this algorithm.

Dead Pixel % Mean	Specifies the threshold level for the determination of dead pixels. Any pixel with intensity below this specified % of mean is deemed a dead pixel.

Defective pixel % Mean	Specifies the threshold for the determination of defective pixels. Any pixel that deviates from the mean dark field intensity greater than specified + or - % of mean is deemed defective.

Maximum Defects per Row or Column	Specifies the maximum allowed defects in a row or column. If a row or column has more than the specified limit the entire row or column is deemed defective.

Maximum Defects in a 5X5 Cluster	Specifies the maximum number of defective pixels in any 5X5 cluster or region.

Maximum Defects of Any Color in a 3X2 Cluster	Specifies the maximum number of defects of any color in a 3X2 cluster or region.

Maximum Number of Dead or Defective Pixels	Specifies the maximum total number of dead or defective pixels in the entire image.

Maximum Number of Defective Rows	Specifies the maximum number of rows allowed in the image.

Maximum Number of Defective Columns	Specifies the maximum number of columns allowed in the image.

Maximum Number of Defective Regions with Cluster Defects	Specifies the maximum number of regions allowed with cluster defects.

Maximum Number of 2X3 Regions with Any Color Cluster Defects	Specifies the maximum number of 2X3 regions with cluster defects of any color.

Background Subtraction On/Off	Specifies if the background subtraction algorithm is enabled.

Background Subtraction Filter Nth Order Kernel Size	Specifies the X and Y grid size of the kernel used for the Nth order filter.

Background Subtraction Filter Nth Order Percentile Threshold	Specifies the percent threshold used to the order classification of the filter.

Light Field Dead and Defective Pixel Outputs

*Name*	*Description*

Dead and Defective Pixel Location Report	A tab delimited text string report output identifies the (X, Y) locations of all identified dead and defective pixels.

Analysis Results Table	A table containing the analysis results for each Bayer plane is output. The number of dead pixels, defective pixels, row defects, column defects and cluster defects are reported.

Bayer plane image	A drop list allows the selection of which Bayer plane will be displayed in the image window. 3D view is also available.

Software