Telit Cinterion, Alif Semiconductor develop a reference design for an intelligent edge camera

Telit Cinterion, a company specializing in IoT solutions, has partnered with Alif Semiconductor, a manufacturer of microcontrollers designed for edge AI applications. Together, they have developed the Vision AppKit, a reference design tailored for intelligent edge cameras. This solution is engineered to perform specific vision AI tasks such as face and object detection, as well as image classification.

The Vision AppKit integrates an Ensemble E3 series microcontroller from Alif and ME310 wireless technology modules from Telif Cinterion. These components work in tandem to capture images and videos, perform them in real-time, and deliver the results on external systems. This solution has a compact design, approximately the size of a postage stamp, making it adaptable for deployment in remote edge locations.

One of the key features of the Ensemble E3 microcontroller is its AI power management technology that dynamically powers only the parts of the logic that are actively in use. According to Telit Cinterion, this intelligent edge camera solution outperforms traditional 32-bit MCUs for AI tasks by at least two orders of magnitude.

Additionally, the ME310 combines power saving mode (PSM) and extended discontinuous reception (eDRX) to ensure the entire system operates efficiently and caters to a wide range of IoT applications. The module supports single-band Wi-Fi 802.11 b/g/n and Bluetooth Low Energy connectivity.

“The Vision AppKit AI camera reference design incorporates highly efficient processing capability to deliver small form factor camera solutions for object detection and image classification in a secure way,” says Reza Kazerounian, co-founder and president at Alif Semiconductor.

Abode Edge Camera for real-time analysis on video feeds

During CES 2024, Abode introduced its edge camera, which integrates Xailient AI computer vision software. This integration enables the system to perform real-time analysis on video feeds. By running AI algorithms directly on the edge camera, Xailient’s technology allows the device to autonomously detect and analyze anomalies in package deliveries or unknown individuals. The company claims that this approach minimizes false positives, which are common in complex systems like this.

In the field of biometric applications, concerns regarding data privacy and security within the IoT ecosystem are on the rise. The implementation of Xailient AI on the edge camera directly addresses this by processing the video feed on the device itself, reducing the necessity to transmit sensitive information to the cloud for analysis. This aligns with the privacy-by-design principle, offering users control over their biometric data.

In terms of wireless connectivity, the Abode edge camera integrates the Wi-Fi HaLow module from Morse Micro. This protocol operates in the sub-1 GHz spectrum, offering long-range transmission and improved penetration through hard objects, such as walls and floors, in contrast to traditional Wi-Fi operating in the 2.4GHz and 5GHz bands.

Furthermore, the Abode edge camera goes beyond motion detection; it includes privacy-safe face recognition capabilities. Users can register family members, enabling the device to differentiate between familiar individuals and strangers. In case an unrecognized person is detected at the door, users receive notifications. These personalized AI capabilities enhance the overall security system, particularly for smart home applications.

Teledyne FLIR machine vision camera for industrial inspection

Teledyne FLIR, a company specializing in industrial-grade imaging solutions, has introduced the Dragonfly S USB3 machine vision camera, designed to meet the diverse requirements of various industries, particularly in vision inspection and quality control systems. This development represents an expansion of the company’s existing embedded camera solutions, aimed at supporting advanced imaging applications.

According to Teledyne FLIR, the Dragonfly S USB3 machine vision camera has a modular architecture, allowing high customization and flexibility for rapid prototyping. This approach enables the user to modify the camera configurations to meet the specific application requirements. Additionally, the Dragonfly S USB3 has a compact and lightweight form factor, which is important for efficient deployment in remote edge locations and integration into at-scale manufacturing processes.

The term “USB3” refers to the interface used in the Dragonfly S camera to ensure high-speed data transfer rates, which is essential for machine vision applications. This feature, when combined with the modular design, enhances the overall system performance, facilitating real-time video and image data processing. The series offers flexibility in the orientation of the USB connection and includes a screw-in locking mechanism for enhanced physical security.

“Dragonfly S has a modular design to accelerate the inception stage of imaging application development for a diverse range of industrial and non-industrial segments,” says Sadiq Panjwani, general manager at Teledyne FLIR IIS. “The modular concept of the new camera empowers engineers to quickly test, develop, and deliver competitive vision products to market at scale.”

e-con multi-camera for high-resolution in automotive and surveillance

German camera manufacturer e-con Systems has developed the e-CAM200_CUOAGX multi-camera solution for the Nvidia Jetson Orin platform, utilizing Onsemi’s AR2020 sensor. This camera is capable of capturing high-resolution 20-megapixel images, with a primary focus on applications demanding detailed image analysis, such as automotive vehicles, surveillance, and robotics.

The camera has the ability to support up to four cameras simultaneously, making it ideal for scenarios that necessitate comprehensive visual coverage from multiple angles and perspectives. This feature is particularly relevant in autonomous vehicular applications, ensuring redundant coverage for safety-critical tasks.

According to e-con Systems, the e-CAM200_CUOAGX can be used for skin analysis, which is important for liveness detection. In the healthcare sector, skin analyzers find utility in obtaining detailed images crucial in dermatology or ophthalmology. Furthermore, the camera’s imaging capability can be leveraged for inventory scanning and digital twin applications.

“e-con Systems has expanded its camera portfolio with an impressive 20MP camera which enhances the capability to capture finer details and deliver exceptional image quality,” says Prabu Kumar, chief technology officer at e-con Systems. “This advancement reinforces our commitment to provide superior multi-camera imaging solutions, ensuring reliability and excellence in every capture for our valued customers.”

Article Topics

Adobe  |  Alif Semiconductor  |  biometrics at the edge  |  edge cameras  |  machine vision  |  Teledyne FLIR  |  Telit Cinterion