Photographs have made a lot of progress in terms of capture, storage, and distribution. The days of photographing with silver bromide film are long gone. These are the days of digital photography, which provides better quality, and better handling, and allows us to easily manage our images.
In this blog, we examine the principles of image sensor technology used in machine vision cameras, and how they are categorized.
Digital Image Sensors
Digital cameras are the device on which the digital image sensor is most frequently used with capturing and storing digital images. The capability of digital imaging sensors has allowed us to take better images than ever before, with many more megapixels. Digital sensors have evolved throughout time, growing in size, resolution, and capability. As a result, we now have greater freedom to take pictures in a way that better satisfies our creative requirements.
Some of the exterior components that are used in the camera will be detailed in an attempt to understand how the sensor is integrated with the camera. The image sensor, which takes the place of traditional film in a digital camera, as well as the lens, which directs light onto it, are its two main parts.
Digital Image Sensors in the data collection
As the cameras continue to improve, so do how we capture images and the ways of using those images in many industries. In the age of artificial intelligence, imaging systems are taking over the role of the human eye. Imaging is one of the fastest-growing modes of data collection and acquisition. A lot of data can be collected using imaging, such as age, gender, and height, from various daily life activities. For example, we can collect data from images when people are standing at a signal, using a vending machine, entering a store, crossing the street, passing a kiosk, registering at some conference, ordering a coffee, boarding a flight, etc.
Imaging not only helps collect information about people, but also about vehicles, packages, animals, birds, fish, medicines, groceries, etc.
Components of Digital Image Sensors
Sensors are categorized according to their structure such as (CCD or CMOS), chroma type (colour or monochrome), and shutter type (global or rolling shutter) etc. They are also classified based on their resolution, frame rate, pixel size, and sensor format. Understanding these phrases will help you decide which sensor is suitable for your application.
Based on the building block of the sensor
Camera Modules are systems that include automotive cameras, image sensors, and lens modules. Image sensors in the camera module transform camera pictures into electrical impulses for processing. For ADAS applications, two types of image sensors are available:
- CCD – Charge-Coupled Device sensors
- CMOS – Complementary Metal-Oxide Semiconductor sensors
CMOS sensors are highly preferred and used due to their low power consumption, ease of integration, faster frame rate, and low manufacturing cost. The detailed comparison between these sensors is explained under the topic of CCD vs CMOS Sensor: Which one is appropriate for Camera Applications?
Based on pixel scanning methodology
Sensors in imaging systems collect and store images for different processing and analysis applications. To snap pictures, these sensors employ an electronic shutter. The exposure of photon wells on the sensor is managed by an electronic shutter. Additionally, it controls whether a full matrix or line-by-line exposure of the pixels is used. The two primary types of electronic shutter are,
- Global Shutter
- Rolling Shutter
The article in the topic of Global Shutter vs Rolling Shutter: Appropriate justification for choosing the best among them described the differentiation of them in detail.
Based on the output format
However, most digital sensors function by capturing light in a grid of photosites, just like a grid of buckets would do for raindrops. Each photosite is exposed to capture incoming light as the exposure commences. When the exposure is finished, the electrical signal representing each photosite’s occupancy is read, quantified, and recorded as a numerical value in an image file. There are two primary types of visible light sensors (but not infrared, ultraviolet, or X-ray);
- Colour Sensor
- Monochrome Sensor
As the most practical solution for image applications, colour cameras have grown in popularity. The truth is that monochrome cameras with trustworthy sensors are a considerably more practical and efficient solution for some embedded vision applications. The rationale is that, when compared to colour cameras, monochrome cameras are able to capture images with excellent details and sensitivity. The topic of Difference between Monochrome Sensor and Colour Sensor helps you comprehend better in understanding of why monochrome cameras have various advantages over colour cameras.
Based on the output interface
The interface for CMOS image sensors is split into two libraries:
- MIPI – Mobile Industry Processor Interface
- DVP – Digital Video Port
The fundamental distinction between DVP and MIPI is that the latter uses a high-speed divergent serial interface, whilst the former uses a parallel interface. In addition to supporting greater resolution and frame rate than DVP interface, MIPI interface offers a wider data bandwidth. Due to the MIPI interface’s complexity and requirement for additional hardware
Image Sensor Format
Image sensors are available in a variety of formats (sometimes referred to as optical classes, sensor sizes, or types), as well as packages. The overall size of a sensor is determined by its resolution and pixel size, with larger sensors typically having better resolutions or larger pixel sizes than smaller sensors. Choosing a lens and other camera optics requires knowledge of the Image Sensor Format. Each lens is created for a particular resolution and format of sensor.
The pixel size is expressed in micrometres (m), which takes into account the full area of the photodiode and any adjacent circuitry. A photodiode, an amplifier, a reset gate, a transfer gate, and a floating diffusion make up a CMOS pixel. Nevertheless, since they can sometimes be shared between pixels, these elements might not necessarily be present in each pixel. It is best to make reference to the sensor’s spectrum response (quantum efficiency) as well as other sensor performance findings to have a more precise knowledge of Image Sensor Sensitivity.
Each one has a distinct function and use case. Not all requirements can be fulfilled by one type. We must select the type of sensor that best suits our requirements based on the use case.
What is the best way to choose a sensor type? What are the advantages and disadvantages of each type? Are there any other subcategories we should be aware of?
Understanding the vocabulary and technology used in digital sensors may help you choose the right camera for your application more effectively if you are familiar with the answers to the questions above. For instance, selecting the appropriate lens will depend on certain sensor parameters, such as pixel size and sensor format. Additionally, you’ll be better prepared to assess if new sensor technologies are advantageous for your application when they become available.
If you are confused with questions and the types, Vadzo’s team will be happy to assist you. Vadzo has the pleasure of working with imaging sensors from businesses such as On Semiconductors, Sony, Omnivision, and others.
Feel free to contact us