After my articles and videos about 3D cameras ' one of the most popular questions was ' 'Which 2D camera should I choose for my project?'
This question is orders of magnitude more complicated than it seems. For most systems that I have seen, 'choosing a camera' is part of the final product ' as important and complicated as 'collecting a dataset' or 'training a neural network.'

If you are looking for more details, kindly visit HBVCAM.

Why?
Today, it's pretty clear that expertise in VLM and Edge models differs. The same is true of cameras. When people develop low-power cameras for home use and cameras for astronomy, they will use different approaches and solutions.

So, I'll warn you right away ' this guide does not pretend to be comprehensive. I tried to focus on a few things:

1. Basic principles of choosing cameras
2. The main types of cameras that account for 95% of the solution
3. Several examples of cameras of each type are good for a start

Also, when preparing the article and video, I asked a few experts to criticize and add their point of view on these issues:

1. Pavel ' CTO Metacomerce. He has been involved in Computer Vision for the last twenty years and has developed camera solutions for the Retail industry for the last five years.
   He proposed the structure of the article.
2. Yuriy ' CEO VIT has been developing traffic control solutions (number plate recognition and everything around) for 25 years, including hardware design.
   He added a lot of information about coaxial connectors and small boards to this article.
3. Andrey ' CEO ORI.team. For the last 5 years, he has been developing industrial systems where IP cameras are actively used for recognition. He also helped with the final polishing of the article.

By the way. This article also has a video on YouTube. In this video I tell a few additional examples and share practice experience:

General logic

When choosing a camera, you need to evaluate several characteristics at a high level:

1. How many frames per second do you need to process? One frame per second is sufficient to issue a parking ticket and 500 fps may be needed to evaluate microsaccades for eyes. This is affected by camera connection type and connection speed.
2. The volume of space that needs to be analyzed, the angle, the resolution, the depth of field, the resolution, etc. Matrix and optics affect this characteristic.
3. Stability to lighting (light, dynamic range, etc.)
4. Where do you need to proceed (the same chip as the camera, in another part of the world, etc.)?

To estimate this, let's check all these characteristics on the camera

Camera connection type

USB, CSI, and LAN are the primary interfaces that account for 95% of cameras. We will talk about them in detail later, and here are a few words about the rest. Partly, these are wrappers over existing ones:

1. PCIe. Many interfaces, like the same USB, are connected to the PCI bus. However, some cameras utilize PCI directly. There are only 2'3 vendors of such cameras. Usually, these are super high-speed cameras.
2. GigE is a protocol over a regular network cable using a UDP channel. It can be considered a subtype of LAN. However, because many cameras specifically support it, it is usually removed separately.
3. CoaXPress (CXP) is a high-speed data transfer standard developed for machine vision and other applications that require high bandwidth and low latency. It is based on a serial communication protocol and uses coaxial cables as a physical medium.
4. GMSL is a protocol used in cars. But it can be a wrapper for the same CSI protocol.
5. HDMI ' nothing prevents you from transmitting/capturing video from the HDMI protocol. For example, monitors can output it via this protocol.
6. FPD-Link ' some cameras have such a connector. Usually, it also uses other existing protocols, such as CSI.
7. Other. Of course, I didn't name everything. There are still analog cameras and Wireless protocols.

Interface speed

Each interface has two main parameters: speed and signal delay. The slowest ones now are old cameras (USB 2.0, etc.). The fastest ones utilize several PCIe lines.

Latency

The further your camera is from your processing unit, the bigger the difference. The more intermediate processing your signal will go through ' the later you will output your result.

Optical scheme

We can talk about optics endlessly. But let's focus on the main characteristics:

1. Focal length (one of the two components of the viewing angle)
2. Aperture size (lens sensitivity)
3. Lens mechanism (zoom, auto focus)
4. Coating (protection from glare)
5. Filters (Polarizing, color, etc.)

Matrix

Main characteristics of the matrix:

1. Matrix size. Number of pixels. Physical size (The larger the pixel, the higher the photosensitiveness and the less noise.) The more pixels you have ' the better resolution you will have. Bigger matrices require a bigger and more expensive lens.
2. Sensor type. This affects how information can be read, the light sensitivity, and the dynamic range. CCD is a reasonably old format, but sometimes it is used, especially if maximum sensitivity is needed. CMOS is now the main format. However, it has a lot of technologies (front light illumination is an old format in which logic and photosensitive elements are printed on one side). Back Light ' appeared about 15'20 years ago. Then, the photosensitive elements are on one side, and the logic is on the other side. Stacked ' a matrix in several layers. For example, each layer catches its wavelength.
3. Spectral characteristics
4. Other parameters (matrix mechanization, active cooling, etc.)

Other ingredients

Also, some cameras have their own light system; sometimes, you need to choose whether to build your system or use existing boxing.

Also, for some applications, it's super important to choose a shutter type. Global vs. Rolling Shutter. Global shutter allows you to have the same moment of exposure for all pixels. Also, for some cameras mechanical shutter, it's still an option(astronomy, for example). But it's quite rare.

Some cameras have an internal CPU/NPU/GPU for image processing, and for some, you need to use another device.

USB

Pros

The first advantage is that preprocessing works on the camera. Some USB cameras can stream a raw image, but most cameras will stream already processed and compressed images. This allows you to utilize the channel width optimally.
The second advantage is that most USB cameras are 'plug and play.' They will work out of the box under Windows andmost Linux systems, MacOS, Android, etc.
The third advantage is availability. A USB camera will be sold in a nearby store. Of course, not every USB camera is so easy to buy, but this will be reasonable for most examples.
The fourth advantage is that USB 3 is a fairly fast protocol. Especially considering compression.

Cons

The first disadvantage is the limited cable length. The camera out of the box will most likely have:

1. The cable is about a meter in general.
2. If the cable is of high quality, then in the best case, up to two meters
3. Cable with active amplification ' about 5 meters.
4. An optical cable can be much longer, but its cost is usually comparable to a good camera's. For example, this cable can reach up to 100 meters and will cost ~ $330.

The second minus. Simple cameras can have a large delay and it can be unstable.

The third minus. Sometimes, finding a ready-made camera with the target characteristics isn't easy.

The fourth minus is that drivers can be unstable. Recently, this part has improved. But 6'7 years ago, the driver could freeze on Jetson, requiring a system reboot.

When to use

In my opinion, USB cameras are super good for prototyping. They allow you to quickly connect something and minimize problems caused by drivers and codecs.

For more information, please visit USB Global Shutter Camera Module.

Very often, small-scale indoor solutions/factory solutions are made on USB cameras, where the camera's price is no longer a determining factor at $200-$300.

USB cameras are good for providing high FPS with high-quality images.
In general, USB is when it is 'Fast' and 'Good' but 'Expensive'.

Examples

Several examples of USB cameras that can be used in projects.
First option. Cheap, quick to buy, good quality for student projects, testing algorithms, etc. ' Logitech C270 (Logitech C920/C922 has a little bigger resolution and framerate)

Second option. Expensive, rich, good quality, lots of implemented interfaces, good drivers and lenses ' Basler cameras. We used them a lot back in '. They where super nice.

Third option. Good Chinese cameras ' Vision Datum. In general, covers most of what Basler can do. But the software is a little worse, the design is a little worse. Etc., etc.

Forth Option. Super Small USB Boards (1, 2). They are nice if you want build your own camera.

CSI

Pros

The most configurable protocol is CSI. You can use a different number of lines or a different approach to transmission. Both the fastest and the slowest cameras are made with CSI protocol.
Cameras transmit raw information ' this allows you to preprocess the image yourself. Also, it's one of the fastest protocols.
There are a lot of camera vendors. You can choose a camera that suits the task.
Also, there are a lot of different connectors based on the top of the CSI protocol.

Cons

Limited length. This can be partially fixed with coaxial connectors, but this increases the system's complexity and cost. People try to stay within the 30cm main range for most everyday applications. But it's hard to find something much longer then 10m.
If your board does not have a separate chip for image encoding, you must do it on the processor. Which will additionally load it
More development time if you go down to the lower level.
CSI cameras are different. They require mutual calibration of the camera and the board. If this is some 'arbitrary camera and arbitrary board,' ' they may not work out of the box.

When to use

When you need to maximize control over the system, when you need to achieve the minimum production cost but do not want to lay out your own matrix yet, or when the final device should be small (this can be achieved with USB, but with CSI, the choice will be greater).
Also, if you are still developing your own board ' using CSI is a fairly logical step.

Examples

Of course, there are a few classic examples that everyone uses
RPi camera ' (works on a lot of boards)

On the Nvidia website, you can find a lot of cameras that work with Jetson.

And here is, for example, another extensive list of different cameras.

LAN

Pros

A clear protocol that will work with all devices. Problems arise much less often than with USB. Lots of goodies out of the box: the camera can be on the other side of the world. Multicast.
There are many cameras on the market. You can find good lenses, good lighting, and competent filters right out of the box.
You can find good cases. Dust and moisture are protected with heating and anti-fogging.
You can find full control: rotation, zoom, pointing, and sharpness.

Cons

High latency ' much higher than USB/CSI. Usually much more expensive than USB and CSI. Cheap models have almost no settings and may not have a good picture.

You are essentially buying a box. No ability to configure internal algorithms. Unless you are developing a camera.
You need a network infrastructure. Most likely, you will need a router and lay networks.

When to use

When latency is not important to you. When there are many users. When you have already installed the camera and the algorithm needs to be run on a ready camera. When you have a central computer where you want to calculate everything. When the working conditions are challenging.

Examples

From expensive cameras, Moxa and Axis are worth mentioning.

Middle class ' HikVision and Dahua.

You can go to Aliexpress for cheap cameras. But remember that for some, even receiving a stream is a problem. The cameras I chose 2'3 years ago are already out of stock. Different cameras from the same vendor may not work, etc.

Summary

There is no straightforward solution to the question, 'What camera should I choose?' Even top companies experiment a lot.
So, the best approach is to schedule an additional amount of time to research this question for your case as well. After this:

1. You write a table of physical and legal requirements.
2. Choose optical scheme + light + power supply.
3. Choose connectivity + protocols.
4. If there are a few possible cameras ' test them/choose the best or the simplest solution.

While you wait for the camera of your dreams, take some simple USB/LAN cameras for experiments.

The company is the world’s best 2mega Pixel USB Cameras supplier. We are your one-stop shop for all needs. Our staff are highly-specialized and will help you find the product you need.

• Do I require a fixed focus or an auto focus camera?
• Does my application need capture at very close distances like 10cm or do I need to capture pictures beyond 20m? How far is my target object from the camera?
• What is the resolution I need to support my application?
• Do I need an OS to use the camera? If yes, what OS should I use?
• Is my environment noisy?
• What type of sensor do I need?
• What is the ambient lighting condition for my application? What is the performance level expected in low-light condition?
• If my application requires video capture, what is the capture rate required at the desired capture resolution?
• If my application requires preview, what is the preview resolution required and preview fps (frames per second) required? How much processing is required after capture of each frame?
• How can I interface e-CAM with my target board? What are the requirements on power supply, clock and interfacing?

• e-con has got ready made drivers for Windows CE / Windows Mobile and Linux for various camera modules on processors like PXA270, PXA320, AU. So if the product that is to be integrated with camera
• e-con has camera solution ranging from 0.3 M to 5M pixel.
• e-con has a good network of sensor and lens assembly makers. In this aspect, e-con would be able to understand the application requirement and would choose the right sensor and lens assembly for the customer.
• e-con has daughter boards for the various camera modules for reference boards like Sirius, Triton starter kit 3 and Triton starter kit 4 . Customers can test their application with various camera modules and once they decide that a particular camera is the right fit for their application then they can go ahead with that. e-con indirectly saves lot of R & D time for the customer in this aspect since tested drivers and daughter modules are available for the various modules.
• e-con has got integrated sample applications with its camera drivers like Video streaming, video capture/still capture and has relationships with third parties which can do 1D/2D bar code capture and facial recognition.
• E-con exposes the API for its driver package with complete documentation and provides sample applications for easy reference
• e-con would act as a one stop shop for the customer regarding the camera.