What Is an Image Processing System?
An image processing system is a series of system configurations that process, synthesize, and read characteristics of 2D and 3D images and data.
Image processing systems have become an indispensable technology for automatic machines and industrial robots because they replace the human eye and enable a variety of judgments and measurements.
Uses of Image Processing Systems
Today, image processing is used in an extremely wide range of fields, including the following:
1. Medical Field
CT and MRI are the two most common types of image processing used in the medical field: CT is a two-dimensional (3D) extension of conventional X-ray images, allowing for the entire body to be observed; MRI uses a strong magnetic field and electromagnetic waves to allow diagnosis without the use of radiation. Both types of examinations use image processing technology to observe the inside of the body from various angles.
2. Industrial Field
In the industrial field, many image processing systems are used in manufacturing lines. They are used for recognition, pickup, and alignment of parts in the assembly process, piece counting, visual inspection, and dimensional inspection in the inspection process, sorting and packaging in the shipping process,. They are also used in a wide range of hazard monitoring, contributing greatly to process automation.
3. Transportation
Typical applications in the transportation field include automobile driver assistance and driving automation. By processing camera images not only from the front but also from the entire 360° angle, the system detects pedestrians, obstacles, and other vehicles, and alerts the driver or takes evasive action.
In addition to automobiles, the system is also used in railroad systems for equipment monitoring and security surveillance, and is useful for monitoring a wide area in a changing brightness environment such as outdoors and along railroad lines on behalf of humans.
4. Security Field
A typical example of use in the security field is facial recognition systems. In addition to its widespread use in smartphones, it is also useful for enhancing security when entering and exiting buildings.
Principle of Image Processing Systems
Image processing systems work as follows:
1. Image Input
Light distribution is converted into electrical signals, mainly using CCD sensors.
2. Smoothing
Smoothing, a type of preprocessing, is a process that gives a smooth out-of-focus gray scale change. Smoothing is also called an averaging filter because it calculates the average value of the pixels in the area covered by the filter and defines that value as the new pixel count. It is used as a spatial filter to smooth an image and remove noise.
3. Feature Extraction
One of the feature images is a binary image. Binarization is the process of reducing an image from several levels of density to only two levels: white and black. An image with only one level of density, either white or black, is said to be a binary image.
One way to determine the nature of an image using tonal values is with a histogram. The number of pixels is taken on the horizontal axis and the frequency of pixels on the vertical axis, and the information is plotted on a graph. The histogram is then processed by dividing the number of shades on the horizontal axis of the histogram into two, with the pixel data being divided into 1 if the number of shades is greater than that, and 0 if the number is less.
4. Evaluation
Feature images obtained by feature extraction are evaluated according to the purpose.
Other Information on Image Processing Systems
1. Camera Selection for Image Processing Systems
Camera selection is very important for image processing. Cameras are used in image processing systems to acquire image data of the workpiece in the image input process.
At production sites, for example, a camera that functions as an eye is used to capture images of inspection objects such as circuit boards in order to inspect the flaws and conditions of products. However, different shooting conditions can cause variations in inspection accuracy.
In order to ensure that the shooting conditions are as identical as possible, the camera must be appropriately selected along with the lens, lighting, and other factors. Image processing systems can be broadly classified into the following two types.
Area Sensor Camera Method
This is the most commonly used imaging method and can obtain a two-dimensional image. The size of the image that can be acquired is determined by the camera.
Line Sensor Camera Method
This method continuously acquires one-dimensional images to obtain two-dimensional images. The camera or workpiece must be moving in a certain direction when acquiring images. This method is effective for capturing images of relatively large workpieces. It is necessary to select an appropriate camera after fully understanding the requirements.
2. Real-Time Processing in Image Processing Systems
Computational processing found in an image processing system, software, or hardware. Software processing is highly flexible because it can be changed to accommodate various changes in the program, but hardware processing is required in situations where real-time processing is required to avoid hazards.
For example, an around-view monitor used to avoid collisions when parking a car originally projects images in real time from above the car, where there is no camera. The image is generated in real time by synthesizing image data from onboard cameras using dedicated hardware, such as an ASIC.