Why do moving objects blur and how to correct this?
Overview
This article will address the cause of blurred moving subjects in video footage and the potential steps to reduce the blurring, along with the possible consequences of making such changes.
Blurred motion: Why does it happen?
Motion blur is a result of the change of location of a moving object. As the object changes location, the light being emitted/reflected from that object changes origin. When a camera views this, the amount of light the camera needs to function will determine the amount of time it processes the image. If the subject moves during this time, the light from the subject will “move” and cause that light to be present on multiple areas of the image, even though it is not. This results in motion blur.
Motion blur is not always bad. TV made considerable use of motion blur to account for the low framerate of TV transmissions. This resulted in what appeared to the eye to be a smooth movement, even though the FPS was quite low.
In a security context, motion blur can cause issues with identifying subjects in the footage. This means that security ideally wants high-speed exposure cameras, but this requirement is at odds with the requirement for cameras to operate with minimal lighting, which requires a longer exposure time. Some post-processing can eliminate this but can result in other issues, such as image noise.
Exposure rate/time: This is not the same as FPS
The exposure time defines how much light a camera will process before generating an image. Exposure rate and time refer to the same concept but are inverse measurements because frequency (rate, measured in Hz) is the inverse of time (sec).
The exposure time for a camera to generate an image increases as resolution increases.
FPS defines when the recorder will create an image for the purposes of display or recording. While FPS can be linked to the exposure rate, it commonly is not, as you usually require a higher exposure rate, than you do FPS. The higher the FPS, the more data required to generate a good quality image, and therefore more storage (recording) or bandwidth (viewing remotely).
As the exposure time increases, the camera will process more light before creating an image. This can be used to account for low-light scenes, where the camera will collect more light before creating an image. This will result in brighter overall image and less noise (detailed below), but the longer exposure means that moving subjects will be less clear as they will have moved during the exposure process.
An example of high exposure time in action, in time-lapse photography. The moving objects are nothing but multiple blurs, while the unmoving public phone box is perfectly clear, exactly because it has remained stationary throughout the exposure. As the exposure time reduces, the camera will process less light before creating an image. This will result in lower overall image brightness, and can cause other issues, such as noise.
Low exposure times are used in activities such as motorsports, where the high speed of subjects would cause considerable motion blur at high exposure times. As a result, the image is a little darker than what you would expect from a full daylight image. Note that the blurriness of the trackside objects is because they are out of focus, not due to motion blur.
Sidenote: Noise
Noise appears as static, similar to old TV transmissions. This static is a result of the electrical signal generated by the collected light being lower than the interference of the electrical power usage in the sensor itself. You will see more noise in darker areas of the image, as there is less overall light (and therefore signal) coming (returning, in the case of cameras with their own lights) from those areas.
Corrective measures and examples
Image examples
These image examples were taken from a 5MP DVR camera. These images are intended to show a contrast between the different settings and indicative of potential results, not necessarily results achieved. This is due to the high number of variables in camera location, lighting and environment.
Auto exposure settings – High motion blur at walking speed
1/25th manual exposure – Motion blur reducing
1/50th manual exposure – Slight fringing, better detail
1/200th manual exposure – extremely short exposure setting, crisp, but the lighting is starting to show as it is insufficient for such a short exposure (Note the bars from the artificial lighting and the slight grey cast to the image from lack of light)
Example controls/settings changes
We will use the NVR-8580 exposure controls in this example, as they are easy to access through the menu. DVR exposure controls are controlled indirectly through the camera image, and not all DVR cameras have exposure controls.
8580 exposure controls
The 8580 exposure controls are located in Display > Image Control in the menu.
Once here, the exposure settings are controlled by the Shutter option. Change the Shutter option to Manual, then the Time Exposure option will appear. Click on the dropdown to see the different time options.
Start with a setting of around 1/25. Experiment with the different options to get the blur down to an acceptable level.
Once the Time Exposure is set to your requirements (check both day and night environments) you can fine-tune the look of the video with the AGC setting (boosts the signal) and the WDR setting (Enable then set the slider). Both settings may increase noise, especially in dark areas of the image. Adjust slowly until the best image is found.
Conclusion
With the rise of extremely high-resolution cameras, camera settings are extremely important to the outcome of the footage captured. Unfortunately, automatic settings are not always the best for every environment a camera will find itself in. Hopefully, this information and examples will assist you in finding the best configuration for your security.