Machine vision is widely used in the industrial sector, covering numerous application scenarios. In manufacturing, tasks performed using machine vision include final inspection of sub-components and checking for potential manufacturing defects in parts. In automation, machine vision plays a crucial role in guiding robots. Furthermore, it is used to verify data matrix codes, inspect food packaging, and read barcodes. This white paper on machine vision provides a comprehensive overview of machine vision systems and the market. This first part will focus on system applications, market trends, and system implementation.
Applications of machine vision systems
Machine vision uses cameras to collect visual data from the environment, and then processes this data through a combination of hardware and software to convert it into a format suitable for various applications. Machine vision technology often employs specialized optical elements to capture images in order to process, analyze, and quantify specific aspects of the images.
These applications can inspect specific characteristics of parts being manufactured on a production line, such as assessing whether a part meets product quality standards, and potentially discarding it if it does not. Machine vision systems use digital sensors housed and protected within industrial cameras equipped with specialized optics. After the sensors capture images, computer hardware and software process, analyze, and measure different attributes of the images to aid in decision-making.
Beyond quality control in manufacturing, machine vision systems have many other applications. For example, they can be used in traffic management systems to monitor and control traffic flow, improving road safety and efficiency. In the medical field, machine vision can support the diagnostic process, detecting diseases early through precise image analysis. In agriculture, machine vision systems can monitor crop health and automate the harvesting process.
Furthermore, in the retail industry, machine vision can facilitate the application of automated checkout systems, improving customer experience and operational efficiency. Therefore, machine vision systems are powerful, widely used, and have become an indispensable part of modern technology.
▷Various visual systems
Machine vision systems are used to inspect a wide variety of products, including food, beverages, pharmaceuticals, and electronics. They employ various machine vision technologies, including image recognition, optical character recognition (OCR), and object detection, to identify and classify defects. Additionally, they can be used to measure the size and shape of objects or to track and trace products throughout the production process. Machine vision systems can help improve product quality, reduce costs, and ensure product safety. By identifying and correcting defects early in the production process, machine vision systems can help avoid recalls and returns.
1. Damage and defects: Machine vision systems can be used to identify various damages and defects, such as holes, gaps, edge damage, pattern defects, bent dipping tubes, damaged or missing seals, and damaged or missing spray triggers.
2. Characters and Codes: Machine vision systems can be used to identify and verify characters and codes, such as dot matrix and non-dot prints, optical character recognition (OCR) and optical character verification (OCV), batches, dates and barcodes, one-dimensional and two-dimensional barcodes (QR codes and custom patterns), as well as for tracking and tracing.
3. Seal integrity inspection: Machine vision systems can be used to inspect the seal integrity of products during seal inspection.
4. Alignment and positioning: Machine vision systems can be used to ensure the correct alignment and positioning of products, such as graphic position and alignment, lid matching with container, etc.
5. Matching and Verification: Machine vision systems can be used to match and verify products, such as ensuring that cap and label patterns match, and ensuring that the spray trigger or cap is facing correctly.
6. Filling and Orientation: Machine vision systems can be used to inspect the filling and orientation of products, such as whether the bottle exists before filling and whether the orientation is correct.
Market information and trends
According to a report by Expert Market Research, the global machine vision market reached approximately US$10.88 billion in 2023. It is projected to grow at a compound annual growth rate (CAGR) of 7.90% between 2024 and 2032, reaching nearly US$21.51 billion by 2032.
The adoption and implementation of advanced innovative technologies such as Industry 4.0 and the Industrial Internet of Things (IIoT) are among the main drivers of growth in the machine vision market. The digital transformation across various industrial sectors, coupled with the accelerated convergence of advanced technologies such as data analytics, IoT, machine learning, cloud computing, and artificial intelligence, is also propelling the application of machine vision technology. To enable real-time decision-making, improve productivity, and increase automation, industries are increasingly focusing on smart factories equipped with computer vision devices, embedded software, advanced sensors, and robots, thus boosting the machine vision market. However, the high initial cost of such advanced equipment and the significant financial and human investment required for frequent maintenance may hinder market growth.
Machine vision applications are gaining popularity due to their low hardware costs, fast and efficient processors, and complete, scalable software that provides all the elements needed to deploy and develop machine vision systems.
The machine vision market in the Asia-Pacific region is booming, driven by rapid growth in automotive, packaging, pharmaceutical, and other industrial applications. As industrialization continues in the Asia-Pacific region, significant increases in investment in advanced technologies across various industrial sectors are expected to create opportunities for market expansion. Furthermore, the robust development of the electronics, semiconductor, and automotive industries in countries such as China, India, South Korea, and Japan will undoubtedly further enhance the optimistic growth prospects of the machine vision market.
Machine vision technology has made remarkable progress since its inception in the 1950s. Its development is closely linked to the evolution of camera sensors. Early CCDs offered high quality but suffered from poor speed and resolution. The advent of CMOS sensors facilitated cost reduction and speed improvements, laying the foundation for the development of megapixel sensors and specialized types of sensors such as infrared and hyperspectral sensors.
The rapid development of deep learning technology has enabled the successful integration of sensors with AI, making real-time object detection and scene understanding possible. With the continued advancement of 3D imaging, edge computing, and advanced sensor technologies, this trend will continue, and future machines are expected to "see" the world with increasingly higher accuracy and intelligence, building an automated and insightful environment for us.
Advances in Artificial Intelligence and Deep Learning
The continued advancements in artificial intelligence (AI) and deep learning have dramatically enhanced the capabilities of machine vision systems. AI algorithms, particularly deep learning techniques such as convolutional neural networks (CNNs), have revolutionized the fields of image processing and pattern recognition. These systems can learn from massive amounts of data and are now able to accurately identify and classify objects, faces, and scenes.
Advances in this area have led to significant breakthroughs in fields such as facial recognition, autonomous vehicles, medical imaging, and security monitoring. As AI continues to evolve, machine vision is expected to become more sophisticated, adaptable, and capable of handling increasingly complex visual tasks. This will undoubtedly drive further development and innovation in the field of machine vision.
System Implementation
In the industrial sector, machine vision is used for electronic component analysis, feature recognition, object and pattern identification, and material inspection. It can help automate various processes and detect faults through image processing. Machine vision is highly popular because it reduces manual labor and improves the precision of product manufacturing.
The system diagram below illustrates how machine vision is implemented for inspecting objects in a factory environment. This section describes the different parts or modules of a machine vision system. Most factory inspection systems use similar modules, but there may be some subtle differences.
Camera Module: The camera module includes a lens and an image sensor, used to capture images of objects for later analysis. The lens selects its focal length and aperture range based on lighting conditions and the characteristics of the object being photographed. The image sensor, located at the image plane behind the lens, is responsible for the photoelectric conversion of information.
Image processor: Utilizes image processing algorithms to analyze digital data from the camera module. The following are the main steps in machine vision image processing:
◆Preprocessing: Preprocessing includes noise reduction and contrast enhancement.
◆Color piping: Color interpolation, color balance, aperture correction
◆Image Recognition:
Segmentation: In this process, a threshold is applied and the edges of the image are determined.
Feature extraction: In this process, size, color, length, shape, or a combination of these features can be extracted.
Processing Unit: A processing unit and built-in software are required to process images and perform detection, measurement, comparison, etc., to confirm whether the images meet quality standards, or to provide type verification or robot control for the system.
Lighting Module: Lighting is one of the most challenging aspects of vision systems. Inadequate lighting or low illuminance on objects or scenes can significantly increase the error rate of a vision system. However, the appropriateness of the lighting applied depends largely on the task at hand.
Sensors: Machine vision systems typically include optical sensors, magnetic sensors, lidar, ultrasonic sensors, and light sensors, forming part of the inspection system. Sensors detect whether there are defects in the final product. Depending on the setup, sensors may also trigger image acquisition and processing, or use some form of actuator to classify, sort, or reject defective parts.
The choice between a monochrome and a color sensor depends on the trade-off between color information and light sensitivity. Color sensors capture natural images through the red, green, and blue channels, making them ideal for applications requiring object identification by color. However, color sensors perform poorly in low-light conditions because the color filter array (CFA) blocks some of the light used to capture color data. In contrast, monochrome sensors forgo the CFA, capturing all incident light in grayscale. Therefore, monochrome sensors are significantly more sensitive in low-light conditions and can potentially process light much faster.
Display: In machine vision, the primary function of the display module is to clearly and intuitively present processed image or video data. In a machine vision system, the display module is a key component used to display and interpret captured image or video data. It presents processed information in an intuitive way, facilitating users to assess data quality, identify potential defects, and make informed decisions based on the analyzed images.
The following factors need to be considered when selecting and setting up a machine vision solution:
In order to select an appropriate machine vision solution, it is necessary to evaluate each stage of machine vision separately.
The purpose of an image capture-machine vision system should be clearly defined from the outset. Take image capture/camera modules as an example. Such systems must be equipped with high-pixel quality cameras with fast frame rates and short exposure times. On the other hand, if product evaluation is required based on its temperature, an infrared camera must be used. In short, the suitability of the equipment depends on the specific application scenario.
Image processing/recognition – Appropriate image processing or recognition software must be selected and integrated into the system used for image capture. The image processing software will run on the hardware, which will determine the image processing speed. The required speed depends on the specific use case; optimizing for the appropriate speed will effectively reduce hardware costs.
System Actions – Image processing and analysis software must be seamlessly integrated with the system taking action. Integration costs need to be factored into the total cost of ownership of a machine vision system.
