FRAMOS Editorial: Why I think Sony got it right when they created the IMX174

FRAMOS Editorial: Why I think Sony got it right when they created the IMX174

A personal recommendation by FRAMOS’ Darren Bessette

There are a lot of sensor choices available now to camera designers and imaging solution providers, available in both CMOS and CCD technologies, with each one having their advantages and weaknesses. Sony is starting to blur the lines between these technologies with the introduction of the IMX174.

Darren_WebsiteA1JNK4zXW0L5lIn case you have not heard about the IMX174, it is a 2.3MP Exmor based global shuttered CMOS sensor (Pregius line) that can run fast, quoting a respectable 164.5 fps at 10 bits and 128.2 fps at 12 bits. The images produced are of HD size and quality at 1920×1200 pixels (16:9 aspect ratio) which matches the size and refresh rates of many of the monitors and TVs sold today. The IMX174 has a larger optical format at 13.4 mm (1/1.2”) with large square pixels, 5.86 µm, bucking the trend of other CMOS sensors used in cell phones which can be found in 1/3” sizes and smaller sporting 8MP resolutions with 1.5 µm pixels. The Exmor technology, with its deeper pixel wells and improved noise performance, make it very sensitive under low light conditions.

The read performance of the IMX174, at < 5 electrons, is better than many CCD based sensors making it ideal for imaging applications that are very sensitive to noise or image summing/stacking common in astrophotography. When taking planetary photographs, amateur and professional astronomers alike are looking for a camera that can capture 1000’s of images quickly with very low noise into one or many files. Then, with some specialized software, they will stack these images to build beautiful portraits of the moon or a planet like Jupiter, where every crater and mountain become visible in high detail. Combine this image acquisition speed with a fully global shutter and you now have a sensor that can really stop moving objects in their tracks without you having to jump through hoops.

Traditional CMOS sensors designed with rolling shutters (Starvis line) are difficult to work with when you are looking at objects that are moving, like cars on a highway. To use these sensors in a traffic/ANPR application, you needed to stop down the aperture on the lens of the camera, set a long exposure value and then use a strobe light to flash the car when the shutter is completely open. In doing this, the camera only sees the car during that brief instant that the flash is going off, effectively stopping its motion in the image. Having a camera with a global shutter eliminates all these headaches and, in some cases, allow you to capture images faster. Now, you just need to set the appropriate aperture to get the right image intensity and fire the flash. The best part is you can do this at the sensor’s full frame rate thus allowing you to take multiple images of the same vehicle with different light sources/conditions to ensure that at least one image has a good view of the license/number plate, the occupants and the color/size of the vehicle.

The large pixel wells provided by this sensor, at about 32,000 electrons, combined with the 12 bits of digitization from the ADC (analog to digital converter) gives a high linear dynamic range of 74dB. In reality though, the signal to noise ratio (SNR) is closer to 44.8 dB which still means that you have almost 7.5 bits of usable image data in each pixel value once you take out the noise. This is really good for a CMOS sensor and is beneficial for applications where you can keep the exposure times low (<1 sec). Once the integration time increases past that, the dark current noise of the sensor starts to become more noticeable and limits the applications in which it can be used. The near IR response of the monochrome sensor is very good and stays high even at higher wavelengths (20% QE at 890nm). This sensor performs well in brighter environments especially if you are concerned about color fidelity. The deeper pixel wells combined with the large bit digitization of the ADC make the IMX174 great to differentiate the subtle hues of color in a captured image.

Let’s face it, Sony does color well!

They understand how to choose the right color pixel filters that do a great job of separating the light hitting the sensor into the primary RGB (red, green, blue) colors. This is welcomed in color quantitative applications like bright field microscopy where technicians and doctors need to differentiate the deep purples from black areas on a haematoxylin and eosin (H&E) stain on a human lung tissue sample. Another area where CMOS sensors tend to be weak is in a high speed, low exposure with low light application. The Sony IMX174 addressed this weakness very well in two ways. The first method is by providing a large gain range in the on-board ADC. The gain implementation is done in two stages, a 0-24dB (1-16x) analog gain and a similar 0-24dB (1-16x) digital gain. With both gains combined, the sensor provides up to 48 dB (1-256x) of gain to really brighten a dark, short exposured image. If that is not enough, you can run the camera in a dual exposure mode. In this mode, you can preprogram two integration times into the sensor and, with each trigger input, the sensor will take two images at two different exposure values automatically. This feature is often used to create HDR (High Dynamic Range) images whereby both images are merged together to provide one HDR image that will show details in both dark and bright areas in one.

Though the overall frame rate is reduced by ½ in his mode, the final images produced rival some of the better CCDs sensors out on the market now. The overall cost of goods of an imaging system using the IMX174 sensor along with this feature will be much less than a comparable CCD based solution while still achieving the same or better data rates. Also, another benefit offered over a comparable CCD is that it can do up to 16 different region of interests (ROI) on the sensor. Most sensors only provide one ROI and, in some cases, there is no speed increase when you reduce the size of the image. In this case, the sensor can be set up to crop multiple sections on the sensor plane and only read out these locations thus speeding up the readout time further. Where this is useful is in factory automation where you have parts going down lanes on a conveyer. There may be distinct locations in the field of view where the part bar code is present so the camera can be configured to only look in these sections thus speeding up the production automation for these parts. I have only listed a few examples of where I see this sensor making a difference. There are many more where it makes sense to integrate it instead of a higher cost CCD.

Don’t expect it to fit in every application but it should definitely be considered in applications where any or all of the following criteria are needed:

  • High speed
  • 2K wide pixel coverage (HD format)
  • Camera and/or target motion
  • High quality color reproduction
  • HDR
  • Price to performance ratio
  • Multiple ROIs

 

If your application has any or all of these requirements, consider if the IMX174 is right for you.  Darren Bessette, Engineering Services Manager at FRAMOS Technologies Inc. Darren Bessette has been creating, developing and supporting vision solutions for more than 15 years. His knowledge and experience has touched all aspects of vision including software and hardware design, image quantification, calibration and processing, optics, and lighting.

Stay Tuned with FRAMOS. Sign Up for our Monthly Newsletters

Contact Us

Salutation*
First Name*
Last Name*
Job Title*
Company*
Email*
Phone*
Country*
What FRAMOS product or service would you like to know more about?*
By submitting the request, your data will be processed to get in contact with you and answer your questions. For more information about the processing of your data, please take a look at our  privacy policy.

Contact Us

Salutation*
First Name*
Last Name*
Job Title*
Company*
Email*
Phone*
Country*
By submitting the request, your data will be processed to get in contact with you and answer your questions. For more information about the processing of your data, please take a look at our  privacy policy.