January 15, 2007

Is Omnivision Technologies Giving up on Wavefront Coding?

Ovtwfc I was looking at Omnivision Technologies presentation at the Ninth Annual Needham Conference and I noticed something very odd in the presentation, there was not a single mention of WaveFront Coding.  In March 2005 Omnivision acquired CDM-Optics for its WaveFront Coding technology also known as extended depth of field (EDF).  Since the acquisition Omnivision has talked about WaveFront Coding and how it will give them an advantage over the competition.  If you look at the older presentations from Jan 2006 forward Omnivision has always presented WaveFront Coding at all the investor conferences.  In article from Seeking Alpha about Omnivision they point out:

WaveFront Coding Technology does not work, and is currently experiencing "serious technology issues."

With the lack of WaveFront Coding in the latest presentation and the article from Seeking Alpha it appears Omnivision has given up on WaveFront Coding.  In my 2007 predictions post I stated that we would not see WaveFront Coding picked up by any major camera phone manufacturer, well it appears one of my predictions is proven to be correct.

Update:

It appears in the 10-Q Filing they seem to be having issues with WaveFront Coding:

During the three months ended October 31, 2006, we shipped the first sample of our new sensor based on our Wavefront Coding(TM) technology. Our customers have indicated a high level of interest in Wavefront Coding, which we are continuing to refine through ongoing internal development efforts. Although continued refinements of the technology will defer the realization of first product revenues, we remain committed to fully exploiting the potential of this advanced lens focus system.

With the recent investor presentation and this filing I can safely say WaveFront Coding appears will never hit the market place in a camera phone.

January 13, 2007

Will Future iPhone's have AutoFocus?

Iphoneb In my previous post I went over what  should be the specs on of the current camera module in the iPHone.  I saw an article published on CNet yesterday titled:

Apple: Calling all iPhone engineers

One of the positions listed caught my eye and that was for an  engineering project manager for its camera features.  From the job listing:

The EPM is familiar with camera module technologies especially sensors, lenses and actuators.

The last part was what caught my eye, they are looking for someone with expertise in actuators forHelimorph their camera phone modules.  The reason this caught my eye is that actuators are being designed now for autofocus and optical zoom in future camera phone designs, one in particular is from Helimorph which is picture on the right.  From the job description, one can assume we should soon see an iPhone possibly with Autofocus capabilities in the next generation.  It really only makes sense that Apple is looking at Autofocus since competitors like Nokia have recently just added it to some of their high end phones like the N93 which has a 3x optical zoom and autofocus.  For Apple's iPhone to be competitive in the camera phone market they really need to improve the image quality from the camera feature.  One personal hope is they don't join the Mega Pixel race, and instead focus on image quality by utilizing technologies like Autofocus and better lenses.  Time will only tell, but personally I think the Apple iPhone will start having a respectable camera that will almost make you want to leave your digital camera at home.

January 11, 2007

What About the Apple iPhone Camera?

IphoneI am sure most of you by now have heard about Apples new iPhone that was revealed by Steve Jobs key note speech at Macworld.  During his keynote, Steve Jobs briefly mentioned that the iPhone will enter the market with a 2Mp camera, but not much more info. Well it got me wondering what kind of performance could we expect and which vendors could be behind the camera in the iPhone.  Late last year Digitimes published an article stating that Taiwanese component makers would be supplying key components to the Apple iPhone.  According the article the following are the camera component makers involved:

  • Largan Precision will provide the Camera Lens
  • Micron will supply the CMOS Imager
  • Altus Technologies will provide the Camera Module Assembly

Well the article is from November and we have to wonder how accurate the details.  Today Information Week released an article talking about which vendors will be supplying the components for the iPhone.  That article verified that Largan and Altus both are involved in the camera assembly.  The only unverified component is the CMOS Imager, who would be the probable supplier.  Currently Micron technology supplies about 40% of the CMOS Imagers used in Camera phones and with their recent acquisition of Avago, that nearly gives them over 45% of the market based on a previous post I did.   Historically Apple has used Micron CMOS Imagers in various products in the iMac, Macbook, and the iSight.  Based on the historical info and Microns current dominance of the CMOS Imager market we can safely assume that Apple has stuck with Micron and is using it to supply the iPhone.

So now with that info what components do we expect to find to make there way into the iPhone.  So I decided to see what camera phone CMOS Imager Micron could potentially supply to Apple, I found that there are 6 potential 2Mp CMOS Imagers.  From the 6 potential four of them were in production and two were future models.  From the remaining options we can safely eliminate the 1/3" models and only focus on the newer 1/4" CMOS Imagers with a pizel size of 2.2µm x 2.2µm can assume that the MT9D112D00STC is going to be the CMOS Imager found in the iPhone.  From Micron's site:

Built with Micron’s exclusive DigitalClarity® technology, this sensor features exceptionally low noise levels and low-light sensitivity. It achieves superior resolution—delivering CCD image quality (based on SNR and low-light sensitivity)—along with the low cost, low power, high performance, small form factor, and fast time-to-market of CMOS.

LarganNow the next thing I wanted to find out is what lens could be used in the iPhone, for that info I went to Largans website and found a list of their current camera phone lens.  From the list the model that closely resembled what appeared to be in the back of the iphone based on the images from Apple Insider is the NB(970) which is pictured to the right.  The specs are as follows: 

  • 4 Element Plastic Lens
  • F# :  F2.8
  • EFL : 3.85(Paraxial)
  • TTL: 5.0

Since Apple has not officially released the full technical specs these are only assumptions based on available data from the known vendors for the iPhone.  The only person to have made any comment on the camera is NY Times David Pogue and here is what he had to say:

I tried out the camera. It was really cool to frame a shot using the HUGE 3.5-inch screen; it’s rare to find that big a screen on any camera. The refresh rate felt typical of a camera-phone to me, but Mr. Jobs said that it would be much smoother by the time the phone is done.

So to summarize what I think the specs of the iPhone will be they are as follows:

  • Micron 1/4" CMOS Imager with a pixel size of 2.2µm x 2.2µm
  • Largan 4 element Plastic Lens

I think we can expect that the iPhone will provide a decent camera that will compare to Nokia and Sony Ericssons current 2Mp Camera phones in image quality.

January 04, 2007

What Will 2007 Hold for Camera Phones?

Well Its finally 2007 and I have decided to list my predictions for what 2007 will hold for camera phones.  In 2006 we started seeing how improved camera phone image quality could play a role in news reporting from the UCLA library incident and the video of Saddam Hussein's hanging; I believe 2007 will have many more videos taken by ordinary people and camera phones.  Also in 2006 we started seeing glimpses of higher mega pixel count and smaller pixel sized CMOS sensors from Micron, Omnivision, and others obviously this will continue and Micron will continue to lead in innovation.  In the CMOS market which I have talked about previously, Micron has made 2006 the year it took the lead from Omnivision and probably will continue to lead in market share in 2007.  The only real threat to Micron in market share is from low cost Asian manufacturers that are just now starting to ramp up there manufacturing efforts, expect 2007 to be the year that asian manufacturers start gaining headway and become a serious player in the CMOS sensor market.  In the optics 2006 we have seen glimpses of real optical zooms coming to camera phones from players like Nokia, Samsung and Sony Ericsson.  This trend will obviously continue and we will see more integrations of some sort of optical zoom.  The technology I think that will make a major impact in camera phone optics this year will be Varioptics liquid lens which I talked about in an another post.  Many of my readers have messaged me and emailed alot is about a technology called Wavefront Coding from Omnivision which was developed by CDM-Optics.  This technology has been in the news for several years and only really appeared to become a real player in the camera phone optics market when CDM-Optics was bought by Omnivision.  Wavefront Coding is CDM's technique for Extended Depth of Field (EDF) which you can learn more about here.  From what I have seen in the press and technical papers I believe that we will not see it deployed in 2007 by any major camera phone manufacturer. Don't count out not seeing EDF coming to camera phones in the future,  companies like DBlur and DXO are working on their own versions of EDF.  So to summarize here are my predictions for 2007:

  1. Camera Phones will become a more dominant in the reporting of news
  2. Micron will continue its dominance in the CMOS Sensor Market
  3. Asian Manufacturers will become a serious player in the market
  4. Varioptics Liquid Lens will finally show up on the market
  5. Wavefront Coding will not be deployed by any major manufacturer this year.
  6. Extended Depth of Field will become an upcoming technology to watch thanks to all the new players in the field.

So with that I hope 2007 will be a great year for camera phone and if you have your own predictions please feel free to email them to me or leave them in the comments.

December 20, 2006

First Look at the Nokia N70

N70_1 I recently recieved a Nokia N70, as a loaner, to review, thanks to the Nokia Nseries N70 Blogger Relations program.  I was very excited to have the opportunity to test and review an N70 camera phone that had a 2mp camera module.  After reading various blogs and review sites about the N70 I had very high expectations on the image quality of the camera.  After recieving the phone I quickly played around with it and took a few sample images of various scenes and I felt that it had fairly good image quality compared to many of the VGA camera phones that I have tested. I played with the various features and was really impressed like always by Nokia's Symbian OS along with the user interface and the installed software packages.  Well when I put the camera in my test bed to begin testing it, this is when my excitement started to turn to disappointment.  To say the least the results were very disappointing and confusing, this had me puzzled.  As I have previously talked about my testing procedure which follows the SMIA specifications. this testing procedure was written by Nokia and ST.  With such surprising results I decided to have a collegue of mine that has extensive years of optics experience to test the N70 and he got near identical results to what I got and was just as surprised as I was.  The image below is the ISO 12233 test target captured by the Nokia N70.

12192006024

I set the various Nokia N70 image parameter settings to auto and to record at highest resolution which is 1600x1200.  I positioned the camera in the best focus distance of the target and to fill the field of view of the camera with the ISO 12233 with 4:3 ratio.  Using the auto timer setting of the N70 I captured the above image.

The CIPA HYRes TestNokian70v

The first test I ran is the Camera & Image Product Alliance (CIPA) HYRes software to measure the resolution of the Nokia N70.  The HYRes software allows me to test the vertical, horizontal, and 45 degree resolutions of the Nokia N70.  In the image to the right is the five line vertical wedge from my test image this allows me to determine the horizontal resolution.  After running this software and obtaining some results this is where I noticed that something appeared to be wrong.   According to  the HYRes software the following are the various resolutions:

  • Vertical Resoltuion is 546 Lines per Picture Height
  • Horizontal Resolution is 490 Lines per Picture Height
  • 45 Degree Resolution is greater then 590 Lines per Picture Height

In my previous testing experience of 2MP CMOS camera modules I expect to see from 900-1200 Lines per Picture Height.  As you see I listed the 45 degree resolution greater then 590, typically when you get this reading you use the 9 line wedges to measure further, but the software could not measure the 9 line 45 degree wedge which is in the upper left side of the test chart.

Spatial Frequency Response (SFR) Test

The next test test I use Imatest to measure the modular transfer function also known as the spatial frequency response.  I have talked about this software and what I measure in a previous review.  This software gives me alot more insight onto whats going on in the camera phone when the picture is taken.  I first ran a test on the lower horizontal bar to determine the vertical resolution of the N70.  In the chart below you can see the results obtained by Imatest.

Nokian70_rb17_lwph

The first graph shows the edge profile of the region of interest in the test image.  The black line represents the luminance (Y) channel.  You can also see faintly red, green, and blue lines; these lines are the respective color channels edge profile.  One of the first things that seems odd is how far off the blue edge is in respect to the other color channels.  The lower graph is the most important, this is the SFR of the Nokia N70.  The black line is the spatial frequency response (MTF) for the luminance Y channel.  Also again you can see faintly red, green, and blue lines; these lines are the respective color channels MTFs.  In my previous review I have explained why I look at the MTF50 reading to determine the vertical image resolution.  The vertical MTF50 is 175.1 LW/PH,  for a 2MP camera this is very low, if you recall in my review of the Motorola Razr which had a VGA camera I measured a 380.4 Line Width per Picture Height (LW/PH).  Based on this information the N70 has the equivalent camera to a sub VGA camera, which doesn't make sense since this is  a 2MP camera phone.  I then ran the same test on the center vertical bar to determine the horizontal resolution and you can see the results below.

Nokian70_ra30_lwph

In this test I obtained an MTF50=308.5 LW/PH for the vertical resolution, in the Razr test I got a 380.4 LW/PH.  This result indicated that obviously the N70 was sub par to a VGA Camera.  So I decided to run further testing to try to find out what could be going on.  Using the 2 far right vertical bars I measured their MTF50 to see how much they varied from the center vertical bar.  In a normal optical system you expect to see the center MTF to have the highest reading and some drop off the further you go away from the center.  In the chart below I tested the far left vertical bar.

Nokian70_rl63_lwph

From the far left vertical edge I obtained an MTF50=166.5 LW/PH, which is nearly half of the center reading of 308.5 LW/PH.  This is a considerable drop off and not typical at all.  I then tested the far right vertical edge and obtained the following results in the figure below.

Nokian70_rr69_lwph

When I ran this test it is when the results got very interesting and started making sense to me what was going on.   You notice that the MTF50=516.2 LW/PH which is nearly 1.5x better than the center MTF.  Based on this reading I would assume that this is a sub 1MP camera not like the previous readings that suggested a sub VGA camera.  This measurement suddenly gave me realization of what is going on and I decided to try one more test on the center and far right vertical bars to solidify what I felt was happening.

Chromatic Aberation

Using the Imatest software I decided to measure the chromatic aberration on the center and far right vertical bars.  Chromatic aberration is one of several aberrations that degrade   lens performance.  It occurs because the index of refraction of glass varies with the   wavelength   of light, i.e., glass bends different colors by different amounts.  In the figure below is the chromatic aberration of the center vertical bar.Nokian70_ra30_ca

You can see how red, green, and blue are so far apart at the peak of the curve.  This indicates that at the center their was signficant effect on the resolution by the chromatic aberration.  In the following figure I looked at the far right vertical bar.

Nokian70_rr69_ca
As you can see in the above figure that the red, green, and blue lines are nearly overlapping at the peak of the chart.  This indicates that their was not much affect of the chromatic aberration at the far right of the image.

Conclusion

Based on the results I obtained, I realized that the Nokia N70 that I was loaned had a serious problem with its camera module.  From the SFR readings of the three vertical bars, I noticed that the camera seemed to be focusing to the right of the image not in the center like it is supposed to be.  With the further test and the results obtained from the chromatic aberration of the center and far right vertical bars I confirmed that the camera was focusing to the right not the center of the picture.  In a typical optical system you would have seen minimal effect from chromatic aberration at the center instead in the N70 I tested it showed signficant effect from the chromatic aberration .   While the vertical bar to the far right displayed the expected chromatic aberration measurement that you would typically find at the center vertical bar.  I am sure many of you are wondering what could cause this Nokia N70 to perform worst than a VGA camera phone?  There are two possible reasons that could cause this to happen.  In my first post I talked about the camera phone module assembly and I divided it into the lens assembly and the CMOS image sensor.  One of the potential reasons that caused this N70 to have such poor results is that the image sensor could be slightly tilted in the module.  The other cause could be related to the lens assembly and that either it is tilted or one of the several elements is off.  I do emphasize that these results are only limited to the Nokia N70 that I have recieved and I am not sure if all of them suffer from these  problems.

Update:

I have contacted Nokia about the issues and was told that they would get back to me after the holidays to see whats going on.

December 17, 2006

The Mega Pixel Myth - Does Higher Pixel Count Necessarily Mean Better Image Quality?

Currently many of today's camera phones have VGA resolution and now we are seeing more 1.3mp camera phones hitting the market with even some 2mp cameras showing up in the higher end units.  One has to wonder will we begin seeing a mega pixel race similar to the digital still camera among camera phone manufacturers.  Personally I hope we don't see a race of who has the most mega pixel in their camera phone at the expense of image quality.  I am sure you are asking why is that, and doesn't having more mega pixels mean better images?  Well David Pogue did a very interesting experiment in New York City for his new show coming in February for the Discovery Channel.  Here is what he did:

On the show, we did a test. We blew up a photograph to 16 x 24 inches at a professional photo lab. One print had 13-megapixel resolution; one had 8; the third had 5. Same exact photo, down-rezzed twice, all three printed at the same poster size. I wanted to hang them all on a wall in Times Square and challenge passersby to see if they could tell the difference.

Well as I am sure many of you will be surprised but almost everyone he asked couldn't tell the difference between the 3 pictures, minus one person who was a photography teacher.  You can read the rest of his article here:

The Truth About Digital Cameras

So you might be wondering how come the camera manufacturers are still pushing higher pixel count and not teaching consumers about these facts, well some are and one of those is Nikon.  Nikon has recently launched a new SLR and instead of pushing some massive megapixel count SLR it is a 6.1 mega pixel camera.  The Economist has done a great job talking about the reasons of why Nikon would make a move like this in the digital still camera market.  According to the Economist :

Nikon is playing the old gamut game—something it mastered while working with photographic film. The gamut of an image is the range of colours in it that can be detected, represented or reproduced in some way. In practical terms, it’s what’s left after an image has been mangled by some output device such as a photographic developer, ink-jet printer, computer monitor, television tube or movie projector.

Basically Nikon knows that in the end its how you plan on using your image if you plan on printing or sharing it via the internet or how ever you wanna display the image.  The image quality is limited by your chosen output device, so as you saw in the previous article from David Pogue most people can't tell the difference between 5mp or 10mp when its printed out as a poster.  You can read the rest of the artice at the Economist:

Nikon's new camera favours quality over quantity

Well since this is a camera phone blog you are probably wondering how this information relates back to camera phones since now most of us only have VGA quality camera phones.  With the trend of new camera phones coming out with higher pixel count I also expect to see a mega pixel race among manufacturers.  As the 2 articles pointed out in the end the more pixels doesn't mean more quality, its really several factors that determine how great your pictures will look, but the most important is how you plan to use that image you capture with your camera phone.

December 16, 2006

Will 2007 Be the Year that We Finally See Liquid Lenses in Camera Phones?

Varioptic has been out for awhile touting their liquid lens technology for quite awhile, here is a brief histoy of how they come about from the Varioptic website:

Our technology is the result of a 10 years research at Université Joseph Fourier and at Ecole Normale Supérieure de Lyon. The founder and inventor of the technology is Bruno Berge, former research scientist at CNRS ( National Center for Scientific Research) and then professor at Ecole Normale Supérieure de Lyon.

Since their beginning they have had a tough journey from a patent battle with Philips and according to that same article issues with their technology having complex manufacturing issues.   Well in August 2006 they overcame many of there manufacturing issues and finally with the help of Creative Sensors Inc. will finally hope to have production line by the end of 2006.  One of their major issues has also been cost, many estimated originally it would cost $5 per lens, to me and you this seems pretty low cost but when current fixed camera phone lens made by companies like Largan which are about $1, they didn't make alot of sense to many camera phone manufacturers.  Well they have now finally realized a price point of $2 and hope by end of next year to hit $1.30 according to the latest article at EETimes:

Camera phone liquid lenses poised for market.

Will we finally see in 2007 a liquid lens in a camera phone?  Well its still to early to predict but with all they have overcome so far to reach this point we might actually see one in some of the higher end camera phones by end of 2007.

December 05, 2006

Are Your Razr Pictures Razor Sharp?

Razr_4 The Motorola Razr is considered the most popular cell phone, and in this post I am going to test the camera on the Motorola Razr.  For the test, I used a Motorola Razr V3 to take the test images.  According to Motorola's technical specifications,  the Razr has VGA camera with 4x digital zoom.  In order to test the image sharpness, I used the ISO 12233 test target to measure the resolution of Motorola Razr.

The Test Image

Razriso_1

Using my camera phone test setup I captured the image above of the ISO 12233.   I set the Motorola Razr light condition setting to auto and to record at highest resolution which is 640x480.  I positioned the camera in the best focus distance of the target and to fill the field of view of the camera with the ISO 12233 with 4:3 ratio.  Using the auto timer setting of Razr, I captured the image above.

The CIPA HYRes TestIsovert

The first test I ran is the Camera & Image Product Alliance (CIPA) HYRes software to measure the resolution of the Motorola Razr.  Using the software I can measure the vertical, horizontal and the 45 degree resolution of the camera.  The software uses the wedges in the ISO 12233 test target in order to do those measurements according to the CIPA Standards.  In the image to the right is the five line vertical wedge from my test image.  According to  the HYRes software the following are the various resolutions:

  • Vertical Resoltuion is 387 Lines per Picture Height
  • Horizontal Resolution is 377 Lines per Picture Height
  • 45 Degree Resolution is 428 Lines per Picture Height

You are probably wondering what the heck those numbers mean, well basically they relate to how sharp the camera is, the higher the line count the sharper the image.  Considering that this is VGA camera these results are pretty decent, an 8 Mp Canon Power Shot Pro1 has a horizontal resolution of 1894 lines so a 377 is not that bad.

Imatest Spatial Frequency Response (SFR) Test

In this test I use Imatest to measure the modular transfer function also known as the spatial frequency response.  So I am sure you are wondering what the heck is SFR well Norman Koren gives a great explaination:

Most of us are familiar with the frequency of sound, which is perceived as pitch and measured in cycles per second, now called Hertz. Audio components— amplifiers, loudspeakers, etc.— are characterized by frequency response curves. MTF is also a frequency response, except that it involves spatial frequency— cycles (line pairs) per distance (millimeters or inches) instead of time. The mathematics is the same. High spatial frequencies correspond to fine image detail. The response of photographic components (film, lenses, scanners, etc.) tends to roll off at high spatial frequencies.

You can read more about MTF and SFR at his site.   Using the Imatest software on the test image I obtained using the Motorola Razr I measure the SFR at the center of the image in the vertical and horizontal direction.  In the chart below you can see the horizontal edge results obtained by Imatest.

Razriso_ya32_cpp_3
The first graph shows the edge profile of the region of interest in the test image.  The black line represents the luminance (Y) channel.  The red dotted line represensents the luminance with Imatest's standardized sharpening applied.  Almost every digital camera sharpens images to some degree, some cameras sharpen images much more than others.  This makes it difficult to compare cameras and determine their intrinsic sharpness.  To solve that issue Imatest develed the standardized sharpening concept to help interpret the actual sharpness so one can compare various digital cameras accurately.  I am not going to spend much time on the edge profile but will come back to this measurement in a later post when I go over the choromatic abberation of the Razr.  The lower graph is the most important, this is the SFR of the Motorola Razr V3 camera.  The black line is the spatial frequency response (MTF) for the luminance Y channel.  The red dotted line is the corrected  spatial frequency respones for the luminance using standardized sharpening.  According to Imatest the best indicators of image sharpness are the spatial frequencies where MTF is 50% of its low frequency value (MTF50).  I measured the Motorola Razr's uncorrected MTF50 is 348.8 Line Widths per Picture Height (LW/PH), while the corrected MTF50 is 357.1 LW/PH.  In the chart below are the results of the vertical edge.   


Razriso_yb16_cpp_1
From the chart above we see the uncorrected MTF50 is 386.1 LW/PH and the corrected MTF50 is 380.4 LW/PH.

Conclusion

Using the HYRes test I was able to to determine the lines per picture height of the Motorola Razr which are:

  • Vertical Resoltuion is 387 Lines per Picture Height
  • Horizontal Resolution is 377 Lines per Picture Height
  • 45 Degree Resolution is 428 Lines per Picture Height

Using Imatest I was able to measure the Spatial Frequency Response (MTF) of the Motorola Razr:

  • Horizontal corrected MTF50 is 357.1 LW/PH
  • Vertical corrected MTF50 is 380.4 LW/PH

Well based on both tests we can see the Razr is fairly decent at taking pictures with large feature sizes and does not fair well with small features sizes.  Using the MTF50 info we can get an idea of how good the image would look if we were to print it out.  Using the following formula: 

   
MTF50(Line Widths / inch on the print) =
    MTF50(LW/PH)      
      
Print height in inches
   
   

and for a standard picture size of 4" x 6" the MTF50(Line Widths / inch on the print) is 89.275.  With an 89.275 that means a 4" x 6" picture will appear adequate but definitely not a "Razr" sharp image.  In the next few posts I will continue looking at various characteristics of the Motorola Razr's camera.

November 30, 2006

Is Nokia becoming the Most Web Friendly Camera Phone Manufacturer?

Last week I talked about which camera phones were the most popular on Flickr, and learned how the Nokia N series was dominating Flickr.  Well it appears Nokia is not only dominating at Flickr, ScanR decided to also release which camera phones were the most popular with their service, the info is in the chart below.  For those of you not familar with ScanR they allow you to use your camera phone to scan, copy, and fax documents. 
Scanr_cameras_1

As you can see in the above chart that the Nokia models dominate the most used camera phones on the site.  You can read more about it at:

Most popular cameras with scanR

November 28, 2006

Camera Phone Testing Procedure - Part 3

In the previous posts, I have covered what targets I use and the setup for taking the images in a controlled environment.  In the first part of this post I will introduce some of the software packages that I utlize in order to do my analysis of the image quality of various camera phones.  In the next post I will cover how I use these tools and some custom tools for measuring image quality with the various targets that I utilze.

Image Analysis Software

The first software I use is called HyRes which was developed by Hideaki Yoshida to accurately gauge the resolution of a digital camera using the ISO 12233 test chart.  The HyRes software uses procedures designed in accordance with the Japanese Camera & Imaging Product Associations (CIPA) standard for  "Resolution Measurement Methods for Digital Camera

Imatestlogo The other software I use is Imatest Pro, which was developed by Norman Koren.  This software is a very powerful image analysis tool that I use to evaluate camera phones' image quality.  This software works with various test targets. Initially, I utilize it with the following targets:  Iso 12233 test chart, the SMIA distortion test target, the Kodak Q-13 target, and the Gretag Macbetch color checker target.  In the next post I will cover all the factors that I test for and how I use each chart to measure that.