Smart Vision Sensor development
Development, Production of the Prototype and Demonstration of the Smart Vision Sensor for Broad Application
Tech Area / Field
- INF-SIG/Sensors and Signal Processing/Information and Communications
- INF-IMA/High-Definition Imaging and Displays/Information and Communications
- INS-DET/Detection Devices/Instrumentation
8 Project completed
Senior Project Manager
Genisaretskaya S V
VNIIEF, Russia, N. Novgorod reg., Sarov
- Excom Inc., USA, NJ, Old Bridge
Project summaryThe project task is the additional applied research performance, development, manufacture and demonstration of an event sensor fundamentally new by its functionality – Smart Vision Sensor (hereinafter – SVS).
The SVS, being as simple in use as a common fire sensor, will be functionally equivalent to a complex up-to-date video system. The SVS is a combination of a video camera with an embedded high-performance specialized processor for video image processing, comparable by its performances with a modern Pentium. General structure of SVS is presented in Figure 1.
Figure 1 – General structure of SVS.
Possibility of developing and embedding into SVS not one, but any required processing algorithm provides extremely wide range of possible technological applications, including, for example, such areas as scene analysis and event detection in the safety/security systems, medical diagnostics, quality control, scientific research, robotics, etc.
The performed information retrieval followed by analysis of about 70 models of known devices referred to as Vision Sensors, GlobalSpec company presents at its website http://video-equipment.globalspec.com/ProductFinder/Video_Imaging_Equipment/Machine_Vision_Inspection_Equipment, has shown that while having bottom price of $1200, no devices are available today close by their capabilities to those planned to be developed as a result of the project activities.
It is quite natural result, because just now the circuit technology has become available that allows SVS development to be carried out at a sufficiently high technical and acceptable commercial level. As the key elements for the SVS development there will serve DSP processor family of Analog Devices Blackfin (www.analog.com), that became commercially available at the end of year 2003 and possess a unique feature set for SVS developing, along with CMOS sensors having attained sufficient perfection. Today at the price of $10 megapixel-sized CMOS sensors are available, in particular, MT9M001sensor of Micron (www.micron.com), practically equal by characteristics to traditional CCD arrays and exceeding them by capabilities of program control.
The proposed SVS in the case of batch production is supposed to have cost not exceeding $200 (without lens and custom-oriented technological software included), power consumption 1W and capability of built-in video data processing comparable with modern PCs. Such device, undoubtedly, should be competitive in the smart video devices market.
The most attractive for us choice of the SVS communication channel type would be USB bus. First, because of its popularity. Second, the USB interface allows it to be used as a port for compact information media (Flash disk) connecting to record alarm frames or some other technological information, what is need for completely autonomous full-featured systems creating on the SVS base.
The project executors present a specialized laboratory having long-term experience of video control and surveillance systems development and deployment, creation of electronic devices for video systems and specialized fast digital image processing algorithms, what is demonstrated by the list of 24 reports and publications./1-24/.
The project activities, duration of which is 3 years, are scheduled so that two results most significant for the commercialization are to be obtained, respectively, by the end of the first and the second year.
By the end of the first year the performed studies and experiments will allow cogent data to be obtained of the device under development technical characteristics and this data propagation to be started by means of presentations and publications aimed at the interested investors attracting.
By the end of the second year there will be finalized activities on the hardware fabrication and software development, conducted in parallel, what will make possible to demonstrate the operating prototype device with two alternative built-in applied functions.
The third year will be dedicated to applied software development and perfection, in particular, basing on the collaborator’s and interested investors’ feedback and recommendations. No less than four new embedded programs will be developed, each destined to solve its own independent technological task.
Thus, the presented Project combines three categories of development in it – applied research, development and demonstration.
Technical Approach and Methodology.
An important moment from the methodological standpoint is the priority purchase of the necessary quantity of engineering evaluation boards for the Blackfin processor. This step will enable us not to postpone efforts on the software development to the moment when the SVS hardware will have been ready. The evaluation boards, supplied with a set of programming and debugging means, will allow us to start algorithms and software developing, both applied and service, practically from the beginning of the project. Software debugging and its efficiency assessment will thus be carried out directly with use of the processor model chosen for SVS creating.
The printed circuit board routing (no less than 4 layers) will be fulfilled by the project executors, and the order for its manufacture will be placed at the specialized enterprise in Moscow. The expected cost of the board at small-batch manufacture is $20. With the number of exemplars in the batch over 200 - $4.
Because of the fact that the algorithms and software development will be carried out beginning from the first months of the project, by the end of the first year, prior to the prototype sensor manufacturing, there will be collected specific data on the future device technical characteristics. This will allow us to begin early proliferation of the obtained results through publications and reports at seminars and conferences, what must positively affect the commercialization possibilities.
Efficiency of the algorithms will be preliminarily tested on a PC, then they will be transferred to a Blackfin in compatible high-level language (С), optimized with the use of the Blackfin assembler within its own programming environment in PC-controlled debug mode, and only after that they will be transferred to ROM or downloadable unit. The code optimization for the Blackfin instruction set and architecture is the most laborious task in process of the applied software implementation. We expect that during this activity the requirements and criteria will be formulated for development of a Blackfin code software optimizer for PC, which could be later on developed either on a commercial basis, or under a separate project.
Scientific and commercial value.
The device to be developed under the project has every reason to become a commercially beneficial, universal device applicable to different problems solving. The essence of our proposal is that such technically unchangeable SVS, subject to the loaded software, will be able to execute special functions depending on the specific area of application. Thus, in industry and technology it could implement remote measurement and control of component shape and dimensions, and the tool condition, as well, in robotics it could serve as a device to provide self-orientation and object recognition, in safety control systems it could survey over the situation stability or admissibility of the changes observed in the scene, etc.
By the expected functionality the proposed SVS will exceed essentially all known products of similar destination, and in the same time, it will be 4-5 times cheaper.
The commercial effect will be formed as a sum of the SVS sale as an electronic device; sales of the applied software solving the requested applied task with the range of technological programs being steadily increased; orders for special technological applications development; program code sale to developers of arbitrary devices on the base of Blackfin-compatible circuit technologies.
It should be mentioned that all SVS software written for Blackfin processor, both service and applied, will possess its own considerable commercial potential, since, as distinct from the PC software market, the Blackfin software market is practically void today.
Expected Results and Their Application.
Thus, upon the project completion following results will be available:
1. Operating prototype SVS.
2. Working documentation that could be applied to arrange SVS batch production.
3. The developed algorithms and programs in Blackfin processor codes providing internal control organization and the service utilities having commercial potential even apart from the SVS.
4. Six programs in Blackfin processor codes, each developed to implement independent applied technology and having commercial potential even apart from the SVS.
5. Hardware core – SVS hardware implementation and programming experience allowing any applied technology developing on customer’s demand.
6. Publications and promotional materials.
The given project meets completely the ISTC goals and objectives, because the proposed development:
· Serves peaceful purposes exclusively;
· The project result possesses a commercial potential;
· Directly employs the scientific potential of the RFNC-VNIIEF laboratory, the staff of which have been previously engaged in nuclear and other scientific war programs, what provides their alternative occupation and involvement into the international scientific community.
The International Science and Technology Center (ISTC) is an intergovernmental organization connecting scientists from Kazakhstan, Armenia, Tajikistan, Kyrgyzstan, and Georgia with their peers and research organizations in the EU, Japan, Republic of Korea, Norway and the United States.
ISTC facilitates international science projects and assists the global scientific and business community to source and engage with CIS and Georgian institutes that develop or possess an excellence of scientific know-how.