Detect smaller features with metrology cameras

Posted by Gretchen Alper on Thu, Jul 3, 2014

As chipmakers continue to move to smaller technology nodes, there are greater demands on process control systems to inspect and measure smaller features and components.

OEM industrial metrology camera detect small features

We talked about future image sensors combining the advantages of CCD and CMOS image sensors that will help to address these challenges in the future:

CMOS versus CCD - what if you don't have to choose


But what can you do if you need to increase performance right now?

One of the simplest ways to detect and measure smaller details is to increase the resolution of the image sensor and subsequent camera.  Previously this had been a challenge because the frame rates of very high-resolution CCD image sensors were way too slow to keep up with the throughput requirements of inline measurement systems, such as 3 fps for 16 Megapixels.  Now there are several new high-resolution cameras available using CMOS image sensors that can produce CCD-like image quality at impressive frame rates, such as 25 Megapixels at 32 fps (or with the ROI function can offer 16 Megapixels at 45 fps for example).

While there are many excellent global shutter CMOS image sensors available, no components are perfect.  The advantages of larger resolution are only realized if the entire image sensor is usable.  To achieve this, it takes an efficient camera design and manufacturing process for defect pixel and blemish correction, accurate sensor alignment, and sensor tuning to maximize the parameters most important for measurement accuracy.

With more than 25 million pixels or even 1 million pixels on an image sensor, not every single pixel can perform perfectly. Uniformity challenges also increase, as a larger optical field-of- view requires more complex optics.  By grading the incoming sensor, dedicated processing and blemish elimination in the camera manufacturing process and camera operation, high-resolution images with high uniformity can be achieved.

Also for high accuracy and precision, particular care has to be given to the optical design and precision of the image sensor placement, especially with high resolution.  If there is any tilt in the mounting and positioning of the sensor, one edge of the image could be sharp while the other edge is blurry making part of the image unusable.  The alignment of the image sensor in the camera is key to have an optimal optical path and high quality image data for the entire resolution.  In the camera design and production process, special attention needs to be paid to the sensor alignment.  Click here for more information on image sensor alignment. 

With the most usable pixels available, the image sensor must also be optimized for registration of the smallest details.  This means the image quality is maximized in terms of dynamic range, sensitivity, image uniformity/spatial uniformity, and linearity.  By tuning the sensor to certain settings, camera manufacturers can reduce the amount of defects the sensor generates. Defects and non-uniformity generation depends not only on the sensor design but also on the conditions it is operated in, like temperature.  Camera embedded calibrations can also be done automatically in the field to adjust to system conditions, such as temperature, optics, and clocking.

Just a quick terminology discussion…while we tend to call any camera designed for industrial applications a “machine vision camera”, this really a broad category.  Machine vision cameras include the low-end industrial cameras used for applications like positioning and mainstream cameras that are designed for general inspection applications such as food and paint quality verification.  Then there are metrology cameras, which are optimized for those applications when the pixel information is used as data for measurements such as in bright field/dark field illumination or interferometry.  Metrology cameras incorporate the specific design and processes as discussed here which enable the quality image data to meet the measurement requirements of smaller technology nodes.

Topics: Vision System Optimization