According to Expert Market Research, the global machine vision market will reach about $10.88 billion in 2023. The market is expected to grow at a compound annual growth rate (CAGR) of 7.90% between 2024 and 2032, reaching a size of nearly $21.51 billion by 2032. The adoption and implementation of advanced innovative technologies such as Industry 4.0 and Industrial Internet of Things (IIoT) is one of the major drivers driving the growth of the machine vision market.

 

In order to advance real-time decision making, improve production efficiency and automation, various industries are increasingly focusing on smart factories equipped with computer vision devices, embedded software, advanced sensors and robots, thus boosting the machine vision market. In the industrial sector, machine vision is used for electronic component analysis, feature recognition, object and pattern recognition, and material inspection, which can help automate different processes and find faults through image processing. Machine vision is popular because of its ability to reduce manual operations and improve the precision of product manufacturing.

 

How to choose a high-performance image sensor? Fully understand the multiple dimensions of performance

 

Key performance features of image sensors include resolution, sensitivity, dynamic range, frame rate and noise level. Resolution determines the detail and sharpness of an image, and high-resolution sensors are suitable for applications that require fine detection, such as medical imaging and quality control. Sensitivity affects performance in low light environments, and high sensitivity sensors are suitable for night vision and security monitoring. Dynamic range determines the performance of sensors in high-contrast scenes and is widely used in traffic monitoring and high-dynamic range imaging. Frame rate affects real-time performance, and high frame rate sensors are suitable for motion detection and automatic driving. Noise levels affect image quality, and low noise sensors are crucial in precision measurement and scientific research. Combining these performance, the selection of image sensor directly determines the effect and reliability of the application, and the performance of image sensor can be evaluated from these dimensions:

 

Resolution: The amount of information per frame or image is the product of the number of horizontal pixels x and the number of vertical pixels y. Unlike consumer cameras, machine vision applications do not require extremely high resolution. High resolution will result in the need for more sensors and greater image processor bandwidth, which will lead to unnecessary increase in system costs. The user must choose the resolution based on the relevant object to be scanned and the amount of light provided on the object.

 

Optical format: Matches the projection of the lens's focused light to the sensor's pixel array to cover the sensor (and make the most of the resolution).

 

Aspect ratio: Whether 1:1, 3:2 or some other ratio, the optimal arrangement should correspond to the layout of the target field of view, so as not to waste the resolution beyond the application requirements.

 

Frame rate: If the target is moving fast, enough images per second need to be taken to "freeze" the motion and be consistent with the physical space being imaged. But like resolution, shooting speed only needs to be able to solve the corresponding problem.

 

Dynamic range (DR) : Factors such as the maximum trap capacity and read noise determine the dynamic range, which is the ratio of the maximum signal to the minimum signal. The larger the dynamic range, the better the sensor is at capturing the gradual details from light to dark in the application scene.

 

Roll-shutter vs. Global shutter: Most current sensors support global shutter, where all pixel rows are exposed at the same time, eliminating blur caused by motion. However, the electronics on the sensor required to achieve a global shutter can add some cost, so it still makes sense to use a roller-shutter sensor for some applications.

 

Pixel size: The larger the physical pixel, the more photons it can accept. In general, large pixels are preferred. However, this requires more silicon area to support the required x, x, y (x by y) array to achieve the corresponding resolution, and requires a larger optical system, resulting in higher bill of material costs. On the other hand, very small pixels require complex optical path design to achieve good optical resolution.

 

Optical sensors can be used for depth perception, orientation and interaction in the environment, and are the only sensor scheme capable of detecting color. It is worth mentioning that image sensors with a global shutter store pixel data throughout the image at the same time without motion artifacts, making them ideal for moving around. Rolling shutter sensors have a higher dynamic range, so they can work better in poor lighting conditions. Global shutters offer advantages in capturing moving scenes and other benefits for developers and manufacturers. Minimizing motion artifacts is a major advantage of global shutters. Global shutter sensors can capture the entire image at once, thus avoiding the distortions and artifacts that can occur in rolling shutter images, especially when capturing fast-moving objects.

 

 

Implement machine vision applications based on Onsemi multiple series image sensors

 

For machine vision applications, the image sensor features such as high resolution, low noise, high frame rate, large dynamic range, low power consumption and high cost performance are very critical. Through its XGS, PYTHON, Hyperlux LP series image sensors, Onsemi meets the needs of industrial machine vision applications for precision, intelligence, and efficiency with the characteristics of high resolution, low noise, high frame rate, large dynamic range, low power consumption and high cost performance, and accelerates the application of machine vision in industrial automation, intelligent manufacturing and other fields.

Thanks to its full-pixel architecture, Onsemi sensors retain more detail in all lighting conditions, even in low light conditions. The scalable product range reduces system development costs and time.

 

The PYTHON family of Global shutter image sensors is a product line optimized for typical machine vision applications, including inspection, tracking, measurement, and more. The series offers eight resolutions based on 4.5um or 4.8um pixels, starting with VGA and up to 25 megapixels. Frame rates range from >800fps to 80 fps, depending on the resolution. All PYTHON devices have a unique "quadratic increment" feature, in which the frame rate increases depending on the vertical and horizontal dimensions of the region of interest being read. PYTHON is available in black and white, color, and near-infrared enhanced versions, as well as a variety of speed and quality level options for selected resolutions.

Hyperlux SG Series The Hyperlux SG is a versatile camera that can be used for a variety of purposes, including AR/VR, machine vision, barcode scanning and device inspection. It has a number of features that are well suited for these applications, including: a global shutter sensor that minimizes motion artifacts, a programmable area of interest (ROI), automatic exposure and control for dynamic lighting, trigger and stroboscopic control for precise active lighting, and low-power operation for battery-powered devices. More sensors with outstanding features will be introduced this year.

 

Onsemi's family of image sensors (also known as XGS) combine high performance, compact pixels and low power consumption. It offers a variety of resolutions, ranging from 2.3 million to 45 million pixels, suitable for a wide range of applications. The sensor also features a global shutter design, which means that all pixels capture light at the same time, reducing distortion and rolling shutter effects. In addition, the sensor is available in black and white and color.

 

Development trend of machine vision technology

 

With the continuous development of industrial automation and intelligent manufacturing, machine vision technology is experiencing unprecedented growth. As the industry's leading image sensor supplier, Onsemi is driving the acceleration of machine vision applications with technological innovation. In the future, image sensor resolution will be further improved to meet the demand for higher image quality and detail capture. At the same time, advances in low power consumption and high dynamic range (HDR) technology will enable image sensors to deliver superior image performance even under harsh lighting conditions, which is critical for applications such as industrial inspection and automated assembly lines. In addition, with the convergence of artificial intelligence (AI) technologies, image sensors will become more intelligent and able to perform more complex image processing and analysis at the edge, thereby improving the overall efficiency and responsiveness of machine vision systems. Onsemi is leading the development of machine vision technology to a more efficient, intelligent and accurate direction through continuous technological innovation.

Previous: Event Vision Technology Unlocks New Vision Of Industrial Manufacturing

Next: TI Introduces New Magnetic Packaging Technology, Reducing The Size Of Power Solutions By Half