[科技新知] The astounding athletic power of quadcopters

Weeks ago, my colleague saw a very interesting video on Ted. He sent me the link and I'm really excited. The reason why he sent me was because we ever discussed the future of the video system month ago. Everyone raised different ideas of the device and my idea is the quadcopters with camera on it.

Why I wanna do such kind of device is oriented from my study in university. I've been studied the mechanics in Aeronautics and Astronautics department, bachelor plus master are total 6 years. There is less such kind of career opportunity in Taiwan because of the environment. Lots of my classmates join the semiconductor industry to be equipment engineers in the company like TSMC, CMO...etc. A little part including me choose different road to be R&D engineer in other industry.

My first resume is to apply a mechanical engineer but my destiny lead me to be a software engineer in surveillance system. Fortunately, I witness the most important period of surveillance. It's from the close system to open system, then open system to smart system. An significant change is the infrastructure of IP network growing rapidly. The network technology overcome the bubble in 2000 and reform the play rules in lots of industry. The open system gradually become mature and we head the future for smart one.

The quadcopter is one example of such kind of devices.
The link: http://www.ted.com/speakers/raffaello_d_andrea.html

I was impressed for the first demo of the inverted pendulum, this is the basic control experiment when in school. Now it was realized on a quadcopter, what amazing! My original thinking is just patrolling from one base station to another but now is far beyond my imagination. Both M2M and H2M are totally implemented in this video. Machines can self-repaired, catch ball, even co-work with each other to finish task and interactive with people. This is a new era of the smart devices not only on smart phone. The time frame from network bubble to now is 10 years and I thought the smart network/device is another 10 years at least. Let's keep our pace on it.

[科技新知] ccHDtv study

What is ccHDtv?

Closed Circuit HD TV (ccHDtv) is a newly developed solution for transmitting full HD digital video in a surveillance system. The core concept of ccHDtv is to deliver high definition video using digital TV (DTV) transmission. With ccHDtv, high definition digital video can be transmitted easily over coaxial cables, twisted pairs, or simply over the air. Multiple 1080p30 and 1080p60 video streams can be easily transmitted using ccHDtv over a single 3C2V/RG59 cable up to a length of 500 meters without any repeater.

In the world of analog TV, The composite video baseband signal (CVBS) is the most commonly used signal format. Its popularity, ease of use, and robustness have made it an integral part of our lives. CVBS is in fact a de facto signal standard for analog video surveillance systems. As we bid farewell to analog TV with worldwide analog switch-off (ASO) and move on to the age of digital TV, digital technologies, such as DVB-T, ISDB-T, and ATSC, became popular and indispensable much like CVBS in the analog age. ccHDtv leverages the extremely robust AV transmission capability of DTV to easily and seamlessly upgrade an analog cctv surveillance system to digital full HD. It is expected ccHDtv to become a de facto standard for digital video surveillance systems in the future.

ccHDtv is a novel technique pioneered by ITE Tech. Inc., ITE has devoted much resource into the development and promotion of ccHDtv solutions. ITE has developed key enabling products, reference system designs, and applications for ccHDtv . These products and applications enable seamless and friendly transition of video surveillance systems from analog to digital, be it all-new digital installations or upgrades from existing analog installations.

Core Technology: COFDM Transmitter / Receiver

More information on wiki: http://en.wikipedia.org/wiki/Orthogonal_frequency-division_multiplexing

Features of ccHDtv

• Reuse existing coaxial cable deployment without any upgrade for cables or connecters.
• Same system planning, deployment and debugging processes as used in traditional CCTV.
• Any iDTV set with DVB-T feature is able to receive ccHDtv signal without an HD DVR.
• Return channels are supported
  • Full duplex communication
  • HD DVR is able to support PTZ control, camera configuration, I/O settings, return audio and remote firmware update.
• Eavesdropping Prevention
  • Support encrypted signals in case of privacy consideration, so only authorized HD DVR can receive and
    decode it.
• No frame drops and real-time delivery
• Quasi-zero latency and quasi-real time preview for 9x, 16x, 32x...
• Long distance transmission
  • 3c2v/RG-59 up to 500M+ w/o repeater
  • 5c2v/RG-6 up to 1000M+ w/o repeater
  • Digital repeater for infinite extender
• High capacity/bandwidth
  • Multiple full HD camera video streams are able to be transmitted over a single coaxial cable.

ccHDtv Full HD Surveillance System

Devices in ccHDtv

Star Topology Deployment

• Exactly the same deployment as the traditional CCTV system
• The coaxial cable and connecter are the same. Actually, the cable medium requirement for DTV CAM is much less than analog cameras

Bus Topology Deployment

• Similar to cable TV deployment (multiple DTV CAM’s running on a single cable)
• Using same coaxial cables, connecters, mixers and amplifiers.
• ccHDtv is very robust and it’s not sensitive to cable network quality variation.
• Ring or multi-path connection is allowed in ccHDtv cable network which enables failure safe feature.

Ring Topology Deployment – Failure-Safe (Redundant) Cable Deployment

• Ring and multi-path connection are allowed in ccHDtv cable network which enables failure-safe (redundancy) feature
• Worst Recover time < 500ms

The above content is from ccHDtv alliance website.

http://www.cchdtv.com.tw/en/intro.php

More FAQ: http://www.cchdtv.com.tw/en/faq/index.php

Specification check points:

• Full duplex communication is supported.(two-way audio and control)
• Multiple stream is support. The max data rate per channel is 32Mbps.
• Wireless transmission is support.(depends on how you understand the regulations. so far ccHDtv suggest not to do so)
• Power on cable is supported. (Both AC/DC power)
• Crypto is support.

Pros compare to HDcctv

• The media is compressed. That means the ccHDtv could have much longer distance than HDcctv.
• ccHDtv supports bus, star topology. The installation is much easier than HDcctv. Besides, the cable cost is also much lower.
• ccHDtv support multiple stream transmission but HDcctv could not.
• ccHDtv support crypto but HDcctv is hard to support.
• ccHDtv use the television standard, the back-end support is much better than HDcctv.

Cons compare to HDcctv.

• None. Maybe the media is compressed.

Pros compare to IP camera

• Ethernet network bandwidth is very limited, and even worse when there are multiple IP CAM’s in the network. To save network bandwidth, IP CAM’s video is encoded with a lower bit rate or less frame rate. Beside, there are frequent frame drops due to overloading bus traffic.
• Each DTV CAM has its own proprietary data channel (up to 32 Mbps), and the guarantee of no frame drop. DTV CAM’s video can be encoded with a higher bit rate and higher frame rate. So the video quality is almost quasi-equivalent to lossless compression mechanism.
• The latency is much stable than IP. What IP can define is the best case.

Cons compare to IP camera

• Theoretically, more than 120 DTV CAMs. Practically, 32~48 cameras are recommended to minimize the adjacent channel (ACI), 2x and 3x inference. That means this solution is much suitable for small/middle projects. However, it could still co-work with IP network by gateway.

The topology show case in the Secutech booth

To make the conclusion:

• We could not see the large replace from original infrastructure to IP system. Only new and big system definitely implement IP system.
• ccHDtv create a much simpler way for traditional VGA system upgrade to HD system.
• ccHDtv can co-existence with IP system.
• IP's NVR system can leverage to support DTV camera. That means, DTV camera is almost an IP camera except the interface.
• ccHDtv use the TV standard and technical. It's robust.

So, in my point of view. This is dangerous to IP camera and it can much slow down the replacement from IP camera.

[科技新知] Panasonic color spliter sensor

There is a new sensor structure that introduce by Panasonic to replace color filter by color splitter. This kind of sensor remove general color filter and add the color deflector before the diode. The conventional color filter would cause 50%~70% light loss and new sensor reduce lots of the light loss. The following pictures show how it works.

The device can also be manufactured using current semiconductor fabrication processes. It doesn't use special materials or processes. It's a great news to bring this sensor to realization.

  

The comparison between the color splitter sensor and color filter sensor.

Panasonic demo with only two-deflector method but various combinations are possible.That means, it could be adjusted by different application.

Although it could provide much better sensitivity, I still have some question about it.

- Would the resolution be sacrificed?

- The diode of different pixel is not balanced, is there any side-effect occurred?

- Would it be much more complicated for the image processing processor supported?

Anyway, I'm still interesting with this sensor and I'll keep eye on it.

[科技新知] Firefox OS - liberate the app stores

I saw a news on TV today that about Firefox OS for mobile phone and it got my curiosity. So I go Mozilla's web site to see what's that about. See the link: http://www.mozilla.org/en-US/firefox/partners/#

The first page have the overview of the Firefox OS. The point to me is following picture. Some of telecommunication operators and device manufacturer get involved in the Firefox OS. It shows that the operators and manufacturer want to liberate the apps stores from the providers like Apple, Google.

Here is a picture that from the Mozilla Taiwan site. It describe the difference the operation mode between original and new Open Web App operation mode.

I think the telecommunication operators would be very interesting with it. Because when more requirements are created for consumers to access the web, the operators can earn more. Developers could have one more option to provide their apps and could dealing with customers directly. For manufacturer, I do not see what they can get except the ODM from operators but they still play an important roles here because variety of devices are required. Would Apple and Google agree this?

[科技新知] Panasonic's 3D CMOS image sensor

3D image/video is blooming in recent years because of the entertainment industry like movie and game. There are also lot kinds of device appear like 3D camera, 3D TV...etc. The player play the 3D image in early stage is by put two camera into one and post processing the two photos into one.

Gopro has the similar structure of their hero camera and we have to buy two camera with some accessories to make this.

There are devices has already embedded two lens and sensors for having 3D functions like Sony camcorder.

I saw a news on Tech-on today that Panasonic developed a 2.1-Mpixel CMOS image sensor for camera systems capable of taking 3D movie with a single lens and announced it at ISSCC 2013. See following links.

http://big5.nikkeibp.com.cn/news/semi/64837-20130222.html

For taking 3D video, it is necessary to separate light into light entering the left eye and light entering the right eye and to make sure that the two types of lights come to different pixels of the CMOS sensor. To realize this, Panasonic employed a structure that combines a lenticular lens and mirror elements (digital micro lenses: DMLs) made by forming patterns whose size is smaller than light wavelength with a lithography technology.

The lights separated by the lenticular lens are collected with the DMLs to increase their intensities. Then, they enter the pixels of the CMOS sensor. With this structure, the light entering the left eye and the light entering the right eye do not interfere with each other much. Therefore, the intensities of the lights become high enough. The received signals obtained in this way are processed via line memory to create 3D video.

Currently, Panasonic is planning to apply the CMOS sensor to industrial and mobile devices, expecting that products equipped with the sensor will debut in 2014.

This may realize the 3D camera system with lower cost because we just need one lens but there are still some information need to be check further. Like the sensitivity, resolution, optical format that lens need, back-end processor and the price of this sensor. I'm very excite for this sensor because it cam simplified the design of the 3D device system. We also can extend the advantages of the 3D system to some applications especially for image system.

[科技新知] Google Glass launch

After I wrote the Vuzix Smart glasses for just hours, Google launched their glass and built a new web site for this product. Maybe google do not want to be left behind when Vuzix award from CES.
http://www.google.com/glass/start/

The page is simple and clear. We could see a video in the link to sense how it feels. All the functions that describe in what it does are totally from audio, video and gesture recognition. We can take the photo, record the video, translation, search the information, navigation...etc in real-time.
All of the recognition technology have to link with the requirements of people and that's why google raise a competition for people to write down "what would you do if you had Glass". Once the idea is adopted, we could become the Glass Explorer and buy Glass with USD $1500.

However, this activity is for 18+ years old and live in the US. Anyway, I could not wait to see what's happen next.

[科技新知] Vuzix Smart Glasses M100

I saw a news that about the Vuzix Smart Glasses M100 and it get a recipient of the CES (Consumer Electronics Show) Innovations 2013 Design and Engineering Award BEST OF INNOVATIONS, and selected as Best Technology in the Wireless Handset Accessory category. More information in following link.
http://www.vuzix.com/consumer/products_browse.html#video-eyewear

It's an interesting device but there are still challenges behind. Since Google launch their project glass, Vuzix is the 2nd company to make the similar product. So M100 get a lot of discussion even google guys go their booth to have it a look. According to the hardware specification from Vuzix web sites. It's totally an android based smart phone and the difference is the glasses appearance only. It's not special by just hardware specification only.

I think the story of this product is based on the intelligence functions like the video recognition, gesture recognition, voice recognition...etc. When these functions are introduced, this device become very interesting.
Vuzix also provide the SDK for developers and it may follow the smart phone pace to have variety of apps on the market.
So, let's look forward to it.