In our last blog post, we discussed the evolving and expanding applications of vision technology, and how significant advances in vision acquisition speed have allowed modern cameras to “see” faster than the human eye, at speeds calculated in thousands of frames per second. But, these advances tell only part of the story. To maximize the capabilities, integrating vision technology with FPGA is required.
With faster and clearer imaging comes the desire to take those images and applying logic or business rules against what the camera “sees”. This actionable data opens opportunities to apply the technology across multiple applications and industries. But, how to leverage a technology that can interpret images as quickly as they appear?
Field programmable gate arrays (FPGA), a technology that has also evolved exponentially, is the perfect pair to advanced camera vision technology to interpret and apply logic to these rapidly captured images. This is not, however, a plug and play application. Significant technical expertise across multiple disciplines is needed to build logic that can interpret and apply algorithms to images appearing at thousands of frames per second.
FPGA-based solutions need to be highly integrated, often consolidating the functionality of several separate circuit boards and additional microprocessor functionality. Expertly development of this type of technology requires a broad range of skills including system design, electrical design, embedded software, and specialized knowledge of programming the FPGA using a hardware description language such as VHDL.
Don’t be fooled — FPGA development is not simply writing VHDL code. VHDL code is merely a software construct to configure your FPGAs hardware. Successful FPGA development requires a team that understands hardware, software, communications, and Real-time Operating Systems (RTOS). Those core competencies ensure that the development results in deep integration of the vision technology with the advanced logic of the FPGA.
A successfully developed and integrated machine vision technology can, in real time, perform the following tasks that were unheard of only years ago.
Configure – Set up the system to acquire the best image possible
Collect – Collect the image data as quickly as the camera creates the image, for example, 5000 fps.
Analyze – Interpret the images per the defined algorithms whether this utilizes visible light, IR, NIR, or other spectral information
Communicate – Display answers, or execute other actions based on the results of the analysis
Integrating FPGA and vision technology is not easy. Resolution’s development team has devoted years optimizing solutions to deliver integrated FPGA and vision technology across multiple industries and business applications. Our core competencies in hardware design, embedded programming, and FPGA development uniquely positions us as the go-to partner for your machine vision development projects.
Learn more about Resolution’s vision integration services, and contact us today to discuss your integrated vision project.