Harvest Imaging Blog

In support of my research group at the Delft University of Technology, I am eagerly seeking for a PhD student (4 years) as well as a postdoc researcher (3 years).  Both persons will be working on CMOS image sensors characterized by ultra-low noise and ultra-high speed.

At this moment the CIS research group at the Delft University of Technology is composed out of 4 PhD students and one postdoc researcher.  But recently a new project is granted and this will allow us to expand the group by two more people.  The job of the two additional people will focus on design, lay-out and evaluation of the CMOS image sensors with the aforementioned characteristics.  The expected background of the candidates is electronic engineering.  Starting date : a.s.a.p.

If you are interested, please send you recent CV to A (dot) J (dot) P (dot) Theuwissen (at) tudelft (dot) nl

(I hope it is clear what is meant here, if I correctly spell the e-mail address I will get tons of SPAM).

Albert, 2-1-2018.

Again another year (almost) has passed by.  And then it is a good moment to take a look back of what happened in 2017.

In the beginning of the year Harvest Imaging announced a project in which the reproducibility, variability and reliability of CMOS imagers will be characterized.  The promise was to deliver a first report in the Summer 2017, but the actual amount of measurement work was a bit underestimated and it took a lot of time to perform all the characterization of the cameras.  New hardware and software had to be installed and as you probably know, these kind of things always take more time than expected.  But early September the first report was published and sent to the customers.  The first feedback of the people who acquired the reports was very encouraging.  More reports will follow in the summers of the coming 4 years.  In the meantime also a dedicated program is started to look for the reliability of the cameras when they are stored at 85^oC and 85% relative humidity.

On the more scientific level we had the International Image Sensor Workshop 2017 in Hiroshima (Japan).  The meeting was again a highlight w.r.t. organization and information content.  In 4 days about 100 papers/posters were presented.  And I think that all participants went home with a happy face.  Not only a lot of high-quality information could be absorbed, but as usually it was great to meet our imaging peers from the academia and the industrial world.  During the IISS-Board meeting, the undersigned was elected as the president of the International Image Sensor Society for the coming 4 years.

The core business of Harvest Imaging is of course the training and courses conducted in the field of digital imaging.  And in 2017 two milestones were reached : course number 200 was organized in Delft.  And of all these courses, 100 of them were organized by CEI.  This “century” milestone was celebrated in Barcelona.  I would like to take the opportunity to thank again all participants who ever attended one or more of my trainings.  Without participants there would be no courses or trainings !!

Finally something about the yearly Harvest Imaging Forum : this year Christian Enz (EPFL) entertained us in the field of “Low Noise for Analog Devices and Circuits”.  Great speaker, and a subject that was, is and will remain a hot topic.  In all my consulting activities I never ever had a customer with the complain : “the noise of my sensor is too low”, it remains the challenge to make today’s noise lower than yesterday’s.

New in 2017 was the start of the Harvest Imaging Newsletter.  Communication to the outside world about the “products” of Harvest Imaging stays crucial.  Very often I do think that the imaging world knows what Harvest Imaging is doing and is not doing, but it happens regularly that people still come with questions about the activities of Harvest Imaging.  So it is a matter of informing and informing my customers again and again about what is when and where available.

So 2017 is almost completed, and then the question is : “What will bring 2018 ?”.  Well I think it will be more of the same.  Although there will be no IISW in 2018, the next one will take place in 2019 somewhere in the USA.  But the trainings/courses will continue (many in-house and open courses are already booked), consulting will continue as well as the work on the reproducibility, variability and reliability project.  Although, it is expected that in 2018 CEI will come with a new product based on my courses.  For sure, I will keep you informed once this new product comes on the market.  Stay tuned !!

Wishing all my readers a Merry Christmas and a Happy New Year.  “See” you soon.

Albert, 12-12-2017.

Together with the speaker/presentater/instructor Christian Enz (EPFL), the final agenda for the Harvest Imaging Forum 2017 on “Noise in Analog Devices and Circuits” is defined.  I am happy to share it with you :

Looks more than appealing !!

All further info about the Harvest Imaging Forum 2017 can be found here.

Albert, 25-11-2017.

This morning, Sept. 25th at 11:30am the first set of CIS devices were stored at 85 deg.C/85% humidity to perform a very first reliability test.  I have no idea what is going to happen to the devices, but my intention is to tease those sensors till they finally will break.  At regular times the devices will be taken out of the climate chamber to fully test their characteristics.

All customers who bought the report “Reproducibility, Variability and Reliability of CMOS Image Sensors” will get automatically get the report on these extra tests.

Albert, 25-09-2017.

There is a tiny small update on the agenda of the 5th Harvest Imaging Forum.  One item is added.

Albert, 21-09-2017.

The 5th Harvest Imaging forum is scheduled for December 2017.  Two sessions were planned, namely on 7/8 Dec. and 11/12 Dec.  2017.  Both sessions will take place !!

I would like to bring to your attention that there will be NO 3rd session for the 2017 forum.  If you still are interested to register, please hurry up before the forum is sold out.

All info about the forum can be found here.

Albert, 16-09-2017.

Actually I am a bit too late to mention that exactly 40 years ago I got my MSc degree from the University of Leuven (Belgium).  As I already indicated in my blog of last year, in 1976 I started to work on my MSc thesis project, and the project finished mid-1977.  The topic of the MSc thesis project was the development of the hardware to drive and read a 256-pixel CCD line sensor.  An object was attached to a small rotating drum and was scanned by means of the CCD line array.  I learned a lot from this practical work, actually not surprising : if you do not know anything of a particular subject, everything you do is new and you constantly learn.

Once I obtained my MSc degree in July 1977, I got the opportunity to stay on-board of the CCD-group and to start a PhD project.  At that time to get funding for their projects, the PhD candidates had to go through a tough oral examination in front of a jury mainly composed out of professors of other (competing) universities.  To prepare myself for this examination, I started working in the CCD-group, headed by prof. Gilbert Declerck. I was involved, in support of another PhD candidate, in the research on the quality of silicon substrates to lower the dark current of the CCDs.  Cor Claeys was my coach during 3 months and he guided me through the clean room.  In this way I got familiar with the silicon processing in general and with the CCD process in particular.  I have seen thousands of silicon defects through the microscope, measured their length and counted the amount of dislocations, stacking faults and other beautiful artifacts.  I got familiar with Wright etch and Secco etch, as well with all kind of annealing steps.  But apparently it was the right preparation for the PhD examination because I could pass the examination, got the funding and could start my own PhD project : the implementation of Indium-Tin-Oxide transparent gate electrodes in the CCD process.  At that time the CCD pixels were based on MOS capacitors with a poly-Si gate electrode.  But the poly-Si did (and still does) absorb part of the incoming light.  To increase the light sensitivity of the pixels, the plan was to replace the poly-Si by a transparent conductive material, such as ITO.  Who could help me with this ?  Actually nobody in our department had any experience with this material and nobody in our department had any experience with the sputtering technique to deposit this material.  Enough challenges ahead for a more-than-interesting CCD-technology project !  The start of a new chapter in my life.

Albert, 13-09-2017.

This afternoon I sent out the very first edition of the Harvest Imaging Newsletter (even with a typo in the title …. so it will become a collector’s item).

Many people have received a copy of the newsletter, but others who are not included on my mailing list will not have received a copy.  For those of you who want to be included in my mailing list for the newsletter, just drop me an e-mail and I will make sure you will get a copy (now and also in the future).

Regards, Albert.

I announced a while ago that Harvest Imaging started a new project on reproducibility, variability and reliability of CMOS image sensors.  This project will last for 5 years, and every year a new report will be compiled that contains all measurement results obtained in this project.  And now the very first report is ready !  It contains 175 pages, 118 figures and 98 tables.  More information about the content of the project can be found at:

www.harvestimaging.com/newproject.php

Find here below the table of content of the first report :

List of Figures                                                                                                                      

List of Tables                                                                                                                              

Introduction                                                                                                                            

Part 1 : Global Shutter CMOS Image Sensor                                                           

          Chapter 1 : Data Collection                                                                                       

                   Conversion Gain                                                                                           

                   Fixed-Pattern Noise in Dark or DSNU                                                     

                   Temporal Noise in Dark                                                                              

                   Fixed-Pattern Noise with Light (B, G, R) or PRNU                                 

                   Quantum Efficiency (B, G, R, n-IR)                                                             

                   Full Well Capacity or Saturation Level                                                            

                   Non-Linearity                                                                                                        

                   Maximum Signal-to-Noise Ratio                                                                     

                   Dynamic Range                                                                                                  

                   Dark Current                                                                                                         

                   Extra Measurements @ 0 s integration time                                                   

                            Fixed-Pattern Noise Analysis                                                                 

                            Temporal Noise Analysis                                                                        

                   Extra Measurements @ 1 s integration time                                                   

                            Fixed-Pattern Noise Analysis                                                                 

                            Temporal Noise Analysis                                                                          

          Chapter 2 : Reproducibility                                                                                      

          Chapter 3 : Variability                                                                                              

          Chapter 4 : Reliability                                                                                                  

Part 2 : Rolling Shutter CMOS Image Sensor                                                                   

          Chapter 5 : Data Collection                                                                                         

                   Conversion Gain                                                                                                 

                   Fixed-Pattern Noise in Dark or DSNU                                                           

                   Temporal Noise in Dark                                                                                   

                   Fixed-Pattern Noise with Light (B, G, R) or PRNU                                   

                   Quantum Efficiency (B, G, R, n-IR)                                                          

                   Full Well Capacity or Saturation Level                                                   

                   Non-Linearity                                                                                                  

                   Maximum Signal-to-Noise Ratio                                                                  

                   Dynamic Range                                                                                              

                   Dark Current                                                                                                   

                   Extra Measurements @ 0 s integration time                                              

                            Fixed-Pattern Noise Analysis                                                               

                            Temporal Noise Analysis                                                                       

                   Extra Measurements @ 1 s integration time                                                 

                            Fixed-Pattern Noise Analysis                                                              

                            Temporal Noise Analysis                                                                       

          Chapter 6 : Reproducibility                                                                                    

          Chapter 7 : Variability                                                                                             

          Chapter 8 : Reliability                                                                                                

Albert, 16-08-2017.

I am happy to announce that the registration for the upcoming Harvest Imaging Forum is open.  Prof. dr. Christian ENZ of EPFL (Switzerland) will entertain us for two days with the topic of NOISE in Analog Devices and Circuits.  More information about the agenda of the forum and about possible registration can be found HERE.

Albert, 06-06-2017.