MACHINE VISION FOR SMARTPHONES

coatiarfΤεχνίτη Νοημοσύνη και Ρομποτική

17 Οκτ 2013 (πριν από 4 χρόνια και 22 μέρες)

63 εμφανίσεις

Essential machine vision camera requirements
to fulfill the needs of our society
MACHINE VISION
FOR
SMARTPHONES
-2-
INTRODUCTION
With changes in our society, there is an increased demand in state-
of-the art smartphones and tablets. This is driving other industries
as well, including machine vision.
The increase in the number of smartphones and tablets requires
production with high-speed inspection with low yield. The
advances in functionality require smaller and more complex
components, resulting in a need for more accurate manufacturing
and measurement. This is all happening on an aggressive time
scale as consumers expect new improvements quickly, resulting in
a fast innovation cycle. Not surprisingly, this is driving innovation in
supporting industries, including machine vision.
High-resolution cameras combined with high speeds that make
full use of select image sensors provide the images required for
inspection and metrology of the latest generation devices that
go inside your latest smart phones and tablets. This includes
supporting the move from
2
D to 3D measurements.
The growth in smartphones and corresponding cameras has
allowed for dramatic improvements in CMOS sensors. This has
also affected the available sensor technology for machine vision
allowing machine vision cameras to then support the production of
more smartphones.
-
3
-
A major trend in society is the needs to be mobile yet
constantly stay connected
1
This has influenced our purchasing
behavior to support how we want to live. For instance, there
has been a dramatic increase in the adoption of smartphones:
“The pattern shows a likely 1 million new Smartphone users
per week being added consistently by the fourth quarter of
this year.”
The ripple effect is dramatic and has resulted in changes and
advances in many markets, including machine vision.
High-resolution cameras combined with high speeds that
make full use of select image sensors provide the images
required for inspection and metrology of the latest generation
devices that go inside your latest smart phones and
tablets. This includes supporting the move from
2
D to 3D
measurements.
The growth in smartphones and corresponding cameras has
allowed for dramatic improvements in CMOS sensors. This
has also affected the available sensor technology for machine
vision allowing machine vision cameras to then support the
production of more smartphones.
While companies like Samsung and Apple are responding to
and feeding the consumers’ demands, the semiconductor
market has grown by focusing on supporting them
3
.
S
MARTPHONE PRODUCTION HAS
INCREASED WITH CHANGES IN SOCIETY
Infographic courtesy AYTM (Ask Your Target Market) and PaidViewpoint
Source: www.asymco.com
2
-4-
I
NCREASED SMARTPHONE GROWTH
REQUIRES HIGH
!
SPEED INSPECTION
WITH LOW YIELD
To increase the capabilities of smart phones, more powerful
processors are required. This means higher density chips,
smaller components and, among others, changes in
packaging.
These changes present new challenges for manufactures of
inspection and metrology equipment. With semiconductor
front-end manufacturing
4
, smaller features must be detected
without compromises in throughput. With semiconductor
back-end manufacturing
5
, there are changes in packaging
such as flip chip
6
technology, which offers significant size
savings. The trend towards continuous miniaturization
results in smaller bump sizes and a greater number of
bumps. This combined with the goal of 100% analysis
at a high precision while maintaining high throughput;
challenges bump inspection and component inspection
equipment manufacturers.
Increase throughput
For all of the inspection and metrology techniques such as
bare wafer metrology or micro defect inspection, precision
and accuracy must be increased while maintaining or
increasing throughput.
Image sensors with a higher frame rate are just a starting
point. The challenge for camera manufacturers is to
preserve image quality at the fastest frame speeds, but it is
possible. The image sensor design must be able to handle
these high data rates within a critical timing tolerance to
reliably produce quality images. The resulting camera
must ensure that the frame rates are actually captured in a
consistent and dependable way. Higher frame rates allow
for gain in throughput. Supporting camera technology/
functionality such as burst mode, CoaXPress and region of
interest (ROI) can further increase speeds.
Throughput can also be increased with fewer movements.
Utilizing high-resolution high-speed image sensors drives
throughput in step-shoot-move inspection systems by
both reducing scan time as well as the number of scan
positions per object. Uniformity challenges increase as a
larger optical field-of- view requires more complex optics
and the increase of defects pixels in the sensor. Camera
manufacturers can provide higher uniformity by grading
the incoming sensor, including dedicated processing and
eliminating blemishes in the manufacturing process and
camera operation.
-
5
-
Improve yield with reliable & stable performance
The goals with better metrology are to detect problems and
defects in order to correct for them to prevent yield problems
or make process improvements to increase yield. Even a 0.1%
improvement in yield can mean an increase in profitability of
millions of dollars.
Bare wafer metrology is a great example of this. Before any
transistor is laid down, the incoming silicon wafer must be
analyzed for flatness and defects. From this inspection , wafers
can be classified to allow the best wafers to be used for the
smallest technology node. Typically measurement techniques
such as interferometry are used for this. Extremely stable
cameras with low noise are needed for accurate measurements.
Cameras with consistent performance reduce the metrology
variability and serves to better determine any process variations.
This reduces process deviations, allowing root cause analysis to
take corrective action. As with all measurements, high quality
means that the variations in the camera and the images are
smaller than the variations of what you are trying to measure in
the production line so you are not measuring within the noise of
the camera.
-
6
-
The push into the
22
and
2
0 nm technology node
6
results
in many changes. There has been an on-going evolution in
front-end semiconductor manufacturing to move from in-line
inspection rather than off-line, destructive analysis whenever
possible. This requires a variety of inspection and metrology
equipment with high quality and stable visible, Infra Red (IR),
or Ultra Violet (UV) sensitive cameras to support the need for
increased accuracy.
The trend towards miniaturization and higher density is of
course carried into PCB manufacturing as well, presenting new
challenges for inspection equipment manufactures to maintain
or improve accuracy. To increase performance while reducing
size results in smaller chips, different packages, higher density
printed circuit board, and multi-layered, more complex boards.
There is also large variety of sizes of the components.
Increased Accuracy
As the objects to inspect/measure with both semiconductor
front-end and back-end become smaller, higher resolution
cameras with better spatial resolution can improve accuracy and
precision. This does require a high quality camera design. As
mentioned before, high quality means that the variations in the
camera and the images are smaller than the variations of what
you are trying to measure so you are not measuring within the
noise of the camera.
Also, particular care has to be given to the optical design and
precision of the image sensor placement in the camera. The
alignment of the image sensor in the camera is key to have an
optimal optical path. This provides the accuracy for the overall
system.
With printed circuit board (PCB) Manufacturing, several
verification steps are required, including automated optical
inspection of the PCBs and components and Solder Paste
Inspection (SPI)
8
. Larger resolution cameras, such as
2
5
Megapixel, can be of benefit when dealing with a huge variety in
components with a very flexible field of view. The move from 2D
to 3D inspection and measurement is another way accuracy is
improved.
With solder paste inspection, 3D inspection and measurement is
becoming more important with changes in the amount of solder
paste used. As the solder bumps and balls become smaller,
the volume of the solder paste is the important measurement
rather than just the width. As the solder provides the connection
between the printed circuit boards, it is critical to measure the
solder volume to verify solder joint reliability. This is done both
pre reflow and post reflow of the solder.
Machine Vision Camera Requirements for 3D
With just a
2
D view from the top, one can only see defects such
as shifts, rotations, and cracks, but not whether components are
flat on the board or the volume of solder paste. With this only 1
image was required to get all of the measurements.
While some 3D measurement systems may use
4
-5 images per
inspected ROI, more advanced systems use
2
0 images or even
more to increase measurement accuracy and to add color vision.
The migration from 1 image for measurement, to multiple
images results in more demands on the camera-based imaging
system. There can be at least two approaches to satisfy these
requirements.
S
MALLER AND
C
OMPLEX
C
OMPONENTS
R
EQUIRE
M
ORE
A
CCURATE
S
YSTEMS
-7-
Option 1 – higher resolution cameras
Higher resolution cameras allow for a larger area to be inspected
at once and provide more data, which can improve accuracy.
BUT since many images are required to perform quantitative
measurements and the overall system throughput must be
maintained, the camera frame rate must also be high. (for
example
4
Megapixel at 180 fps or even
2
5 Megapixel at 3
2
fps
and higher)
Since multiple images are combined, the stability and
reproducibility in the camera is more critical than in the past.
Only intentional changes can occur between the images. This
means black level, gain, among others must be exactly the same
for all of the images. The camera manufacturer controls all these
parameters through careful design and implementation.

Option 2 – multiple cameras
Another option to reach these goals is through multiple cameras
to capture all of the images. This could mean fewer illuminators
and less stringent requirement on the speeds of the camera.
This is attractive as it allows for scalability using more cameras
for higher end systems, and can seem more cost effective since
“lower-end” cameras can be used. This should be done with
caution though as the cameras can have lower frame speed,
but need to be extremely consistent, and well-matched for this
technique to be accurate.
-
8
-
Why is it that cameras for consumer electronic products, e.g.
smartphones have more than 5 megapixel tiny cameras that cost
next to nothing, are not used for machine vision?
The larger pixel image sensors (greater than 5.5 um) can allow
for the best accuracy (i.e. Full Well and Read Noise), but they
also result in the highest costs due to large sensor sizes (silicon
real estate consumed) and additionally expensive optics. Larger
pixels are still used in the industrial and scientific market, but
the trend in other markets has been towards much smaller
pixel sizes. This is especially so with CMOS image sensors.
These new image sensors are enabling better cameras for our
smartphones and web cameras, with pixel sizes down to 1.4
um and extremely low cost. For machine vision, CMOS sensors
with smaller pixels (even
2
to 3 um) may not be acceptable,
especially with high-end inspection applications, such as
semiconductor inspection, Flat Panel Display inspection, or
electronic metrology applications. Smaller pixel image sensors
should reduce the cost of the camera because of the camera
size, or more pixels inside the same camera and optics, leading
to higher resolution.
These benefits are all appealing for machine vision too so what
is given up? Our conclusion, based on a thorough analysis is
that with pixels less than
4
.5 um, is that too much functionality
and performance is sacrificed for a lot of machine vision
applications
9.
That being said, the constant drive for innovation with
smartphone cameras has led to dramatic improvements in
CMOS image sensors that are well-suited for machine vision
cameras in a short time frame. These latest global shutter
CMOS sensors allow for the machine vision cameras with high-
resolution combined with high-speed that are required to meet
the accuracy and throughput needs of the latest inspection and
metrology equipment as mentioned above.
The growth in smartphones and corresponding cameras has
allowed for dramatic improvements in CMOS sensors. This has
also affected the available sensor technology for machine vision
allowing machine vision cameras to then support the production
of more smartphones.
C
ONCLUSION
Our increasing need to stay mobile and connected translates to
a worldwide adoption of smartphones and tablets. This trend
has a major impact on the speed and growth of innovation. In
the end, every OEM that uses machine vision benefits from this,
including food inspection equipment or even intelligent traffic
systems.
While better CMOS sensors are allowing us to take better
photos with our smartphones, better CMOS sensors are also
used in the industrial cameras that guarantee the quality of
the parts within the phones. High performance machine vision
suppliers have been relied on to enable the advancements
required. This includes the move from
2
D to 3D.
I
NNOVATION
D
EMANDS
D
RIVE
F
AST
I
NNOVATION
-
9
-
REFERENCES
1.

http://www.peoriamagazines.com/ibi/2012/jan/year-of-mobility
2
.

http://www.asymco.com/
2
01
2
/03/0
7
/the-unrelenting-trends-in-the-us-smartphone-market/
3.

http://www.electroiq.com/semiconductors/
2
01
2
/0
4
/19/tablet-and-smartphone-sales-driving-


global-semiconductor-market.html
4
.

http://www.adimec.com/en/Service_Menu/Markets/Machine_vision_cameras_for_semiconductor_


wafer_metrology
5.

http://www.adimec.com/en/Service_Menu/Markets/Cameras_for_back_end_semiconductor_


packaging_inspection
6.

http://en.wikipedia.org/wiki/Flip_chip
7
.

http://en.wikipedia.org/wiki/
22
_nanometer
8.

http://en.wikipedia.org/wiki/Automated_optical_inspection
9.

http://info.adimec.com/blogposts/bid/6868
4
/Can-small-pixel-CMOS-image-sensors-be-useful-in-


Machine-Vision
-10-
Adimec specializes in the development and
manufacturing of high-performance cameras that meet
the application-specific requirements of key market
segments, including machine vision, medical imaging,
and outdoor imaging. Founded in 199
2
, the company
partners with major OEMs around the world to facilitate
the creation of industry-leading cameras.
The unique Adimec True Accurate Imaging
®
technology
provides new levels of precision and accuracy to vision
systems. Its diverse line of camera products meet a
wide range of performance, size, cost, interface and
application requirements. Adimec has offices around
the world focused on creating customer value and
satisfaction through local, personalized support.
Need more inspiration? Contact us www.adimec.com
ADIMEC