Embedded vision systems have seen rapid development in recent years in terms of their performance and applicability to a wider variety of use cases. Many modern-day embedded vision systems utilize multiple types of lenses and sensors that come with different feature sets and at varying costs. The design of the camera systems integrated with these components plays a vital role in achieving the required performance and image quality.
However, the selection and evaluation of sensors and lenses are easier said than done. The right combination of both can help you build a highly optimized embedded vision system that meets all your quality and performance standards.
When it comes to selecting a lens for an embedded camera, there are numerous factors to be considered. And one among them is FOV or Field Of View.
In this article, let us dive deep into the importance of FOV in embedded vision applications. We will also have a detailed look at the factors that determine the FOV of a camera system in addition to learning the importance of FOV in embedded vision.
What is Field of View?
Field of view (FOV) is the maximum area of a scene that a camera can focus on/capture. It is represented in degrees. There are three ways to measure the field of view of a camera – horizontally, vertically, or diagonally as shown below.
Figure 1: HFOV, VFOV and DFOV
Generally for a sensor, FOV refers to the diagonal measurement – which is called DFOV or Diagonal FOV. Horizontal FOV (HFOV) and Vertical FOV (VFOV) will vary based on the aspect ratio of the image sensor used.
How to calculate FOV?
Now let us discuss FOV calculation. In many applications, the required distance from an object and the desired field of view (which determine the size of the object seen in the frame) are known quantities. This information can be used to directly determine the required angular field of view (AFOV) as shown below.
Figure 2: AFOV (Angular Field Of View)
The formula for calculating AFOV is shown below.
Where,
AFOV – Angular Field of View
HFOV – Width of the object seen through the lens
Working distance (WD) – Distance between the lens and the object
Conversely, if you know the FOV and the working distance, then you can calculate the dimension of the object using the below formula.
Similarly, for the calculation of VFOV and DFOV, instead of width (or horizontal F0V), corresponding height and diagonal dimensions of the object are substituted in the above formula respectively.
For example, imagine that the camera and the object are fixed at a working distance of 30cm. In this case, the HFOV and VFOV are measured manually using a scale (in mm) as shown below:
Figure 3: Manual HFOV and VFOV calculation
Then the values in mm are converted to degrees using the above formula.
The importance of FOV in Embedded Vision Applications
FOV is one of the most critical parameters considered while integrating a camera into an embedded vision system. Whether it’s an intelligent transportation system, autonomous mobile robot, remote patient monitoring system, or automated sports broadcasting device, FOV plays a major role in ensuring the necessary details of the scene are captured. The FOV of the lens can be set as wide or narrow based on the end application requirements.
To learn more about lens selection, please visit How to choose the right lens for your embedded camera application.
Factors that determine the FOV of a lens and embedded camera system
The major factors that decide the field of value of a camera are focal length, sensor size, and working distance. Let us look at each of them in detail.
Relation between Focal Length and FOV
Focal length is the defining property of a lens. Simply put, it is the distance between the lens and the plane of the sensor, and is determined when the lens focuses the object at infinity. It is represented in mm. Focal length depends on the curvature of the lens and its material. The shorter the focal length, the wider the AFOV and vice versa. Please have a look at the below image to understand this better:
Figure 4: Focal Length and AFOV
The below equation relates the angular field of view of a lens with the focal length.
Relation between Sensor Size and FOV
Each embedded vision application has different sensor size requirements to get the best output. A small sensor will have a narrow field of view while a large sensor can provide a wide field of view.
You also have an option to capture the same field of view with sensors of different size. This can be done using a lens with the appropriate focal length. As a result, the same FOV can be achieved using a small sensor with a short focal length lens and a large sensor with a long focal length lens.
The below image illustrates the concept:
Figure 5: FOV Result with small and large sensors (Source: Panavision)
Relation between Working Distance and FOV
FOV also depends on the distance between the camera and the object. As discussed earlier, if the objects are closer to the camera, the FOV becomes wider. This is because shorter focal lengths require shorter working distances for proper focusing. Thus, the lens to sensor distance has to be designed based on the working distance.
How to choose the right Field Of View for your application?
From the previous section, we understood the definition of FOV and its relation with several other lens parameters. Let us now discuss how to choose the right FOV for an embedded vision application.
Most of the embedded camera applications require the FOV to be wider enough to cover a large viewing area. For instance, a fish-eye lens is characterized by wider FOV and larger depth of field (DOF) and hence is suitable for surveillance applications. On the other hand, for a zoom/telescopic application, you might require a normal/narrow FOV.
Also, let us consider a popular embedded vision application like autonomous mobile robots (AMR). These autonomous systems perform obstacle detection and obstacle avoidance (ODOA) to seamlessly navigate their environment. And many of these robots require FOVs in excess of 180 degrees. This ultra-wide FOV is achieved by using multi-camera systems.
Having 2 or more cameras enables a higher resolution, prevents lens distortion, and offers a wider FOV. To achieve high imaging quality in multi-camera systems, a lens having an FOV of around 60-70 degrees is usually chosen. But it is important to note that this is determined by a multitude of factors. There is no ‘one-size fits all’ approach to this. It is recommended to take the help from an imaging expert like e-con Systems as you go about picking the right field of view and lens for your application. Please feel free to write to us at camerasolutions@e-consystems.com if you need a helping hand.
Meanwhile, you could check out the article What are the crucial factors to consider while integrating multi-camera solutions? if you are interested in learning more about multi-camera integration.
e-con’s Wide FOV Camera – e-CAM130A_CUXVR_3H02R1
e-con Systems has led from the front when it comes to innovation in embedded vision. And one of our key strengths is the platform side expertise especially on the NVIDIA Jetson series. Leveraging this, e-con has designed many multi-camera solutions that offer an FOV of up to 360 degrees.
And one of the most popular among those solutions is e-CAM130A_CUXVR_3H02R1 180° FOV camera – a synchronized multi-camera solution that can be directly interfaced with the NVIDIA® Jetson AGX Xavier™ development kit. This camera solution comprises of three 13 MP camera modules that are based on the 1/3.2″ AR1335 CMOS image sensor from onsemi®. These 4K camera modules are positioned inwards to create a 180° FOV as shown in the image below:
Figure 6: e-CAM130A_CUXVR_3H02R1 setup
What helps e-con stand out when it comes to this solution is our proprietary 180-degree stitching algorithm that can process images from multiple cameras to create a 180 degree image. To learn more about this solution, please visit the product page.
How e-con helps customers choose a lens with the right FOV
e-con has its cameras integrated into 300+ customer products that are in the market currently. And with the rich experience of working with hundreds of clients, we understand the nuances involved in the selection of lenses for a camera application. Whether it’s about the field of view or depth of field, we can guide you through the entire process of selecting the lens rather than merely acting as a camera supplier. We also offer extensive customization services on the camera side. Please do check out the OEM camera customization page to learn more about our full suite of customization services.
Also, you can visit our FOV calculator page to easily calculate the FOV required for your lens based on your application.
If you are looking for help in integrating cameras into your products, please write to us at camerasolutions@e-consystems.com. In case you are looking for the right camera for your application, the Camera Selector is where you can find our entire portfolio of products.
Prabu is the Chief Technology Officer and Head of Camera Products at e-con Systems, and comes with a rich experience of more than 15 years in the embedded vision space. He brings to the table a deep knowledge in USB cameras, embedded vision cameras, vision algorithms and FPGAs. He has built 50+ camera solutions spanning various domains such as medical, industrial, agriculture, retail, biometrics, and more. He also comes with expertise in device driver development and BSP development. Currently, Prabu’s focus is to build smart camera solutions that power new age AI based applications.