If the shutter speed on your camera is fast enough, it is possible to have ghosting without motion blur in your image. If the slowest shutter speed on your camera is 1/500th of a second, it will freeze most moving objects; therefore, you will not necessarily have motion blur in the image. However, regardless of how fast your camera’s frame rate is, there will still be a quick duration of time in which moving objects have time to move throughout each frame, thus causing ghosting. Therefore, you will still have ghosting, but you may not capture motion blur in your image.

In a camera system, the image sensor receives incident light (photons) – either focused through a lens or any other optics. Hence, lens selection plays a major role in determining image quality, FoV (Field of View), DoF (Depth of Field), etc.

For instance, if you are using See3CAM_CU135 – a 13 MP high-resolution USB camera from e-con Systems™ – to read barcodes, you need to make sure that the lens you pick can meet the maximum desired resolution.

In today’s blog, let’s look at how to determine the resolving power of a lens, which is one of the most critical parameters to consider while choosing a lens for your application.

The resolving power of a lens can be calculated manually as well as in automated manner. The manual method is called the human eye perception method, and the automated technique is called the IMA Test method. We will look at both these in detail in this section.

If you are looking for help in selecting a camera solution with the best-fit lens for your application – no matter the industry, please write to us at camerasolutions@e-consystems.com. You can also visit our Camera Selector to get a full view of e-con Systems’ camera portfolio.

To understand how the theoretical and practical methods of calculating resolving power differ, we will look at how these are done for e-con Systems’ See3CAM_CU135 in the next section.

ImageComics

Resolving power is calculated as object space resolution. And object space resolution is derived from what is called the image space resolution. We will now look at both the terms and learn now they are calculated.

Using the above data, you can validate how much lp/mm a particular lens would resolve at a specific working distance. Based on this, you can select the right lens for your application. It is pertinent to note that a change in FoV will affect the magnification factor, which will end up affecting the practical image space resolution.

For instance, if you want to calculate the resolving power of See3CAM_CU135 – 4K USB camera – with the default product lens, you should capture the image of the resolution chart with the desired aspect ratio at the prescribed working distance.

GloryImageComics

Let’s say you have to decode an object like a small barcode and that too at a longer distance. You would have to differentiate the bars by allowing a certain amount of space between them. This minimum distance depends on the resolution of the camera. However, when you use a sensor with the desired resolution, it is also important to choose a lens that can help capture the level of detail expected from that resolution. This is where knowing the resolving capacity of the lens becomes important.

The reason why this is called the human eye perception method is that the number of lines counted may differ depending on the ability of the observer to distinguish between two consecutive lines.

A line pair is a pair of black and white lines next to each other with the same width and orientation. The ability to differentiate two bars as separate entities in a specific resolution would be based on the contrast level. It means that calculating resolution in terms of lp/mm is extremely useful when comparing lenses. This can act as one of the criteria while choosing the best-fit lens for a given sensor and application.

GhostedCompendium

As mentioned above, Object Space Resolution denotes the resolving power of a lens. It defines the size elements of the object that can be resolved. It is calculated as:

The resolving power of a lens is measured in line pair per millimeter or lp/mm. It is a measurement of spatial resolution used to calculate how small a detail in an image can be resolved by a lens. The unit expresses the number of line pairs you can fit within one millimeter.

Given below is the comparison of image space resolution and object space resolution values obtained using the theoretical and practical methods.

Image Space Resolution is the resolution in the image plane in consideration of the sensor pixel size. Generally, two pixels or one line pair is the highest frequency which can be resolved by a sensor – using the Nyquist frequency. Therefore, image space resolution is theoretically calculated as:

Ghostedcomic

It is also possible to get a combination of motion blur and ghosting in your image. Unlike ghosting where you can have one artifact (ghosting) without the other (motion blur), if you have motion blur in your image, you will have ghosting as well. For example, if there are moving objects in your scene and your shutter speed is slow, you will see a combination of motion blur and ghosting. Just remember that, depending on the overall shutter speed settings, ghosting can come without motion blur but motion blur always comes with ghosting. Now, there are many photographers who intentionally use motion blur and ghosting in their HDR images to create interesting effects, so ghosting and motion blur are not always bad.

To give you an example, if the aspect ratio considered for See3CAM_CU135 – 13MP USB camera – is 4:3, the image of the resolution chart has to be captured at that aspect ratio to finally arrive at the resolving power. Below is a sample image of the resolution chart taken at an aspect ratio of 4:3.

In the image below, the moving cars look a little transparent and faded, hence the name “ghosting.” There is also motion blur in this image, which we will explain next.

The other type of artifact that we will get in HDR photography is motion blur. Motion blur is not necessarily a factor of the HDR process but rather the slow shutter speed on a camera. Motion blur occurs when the shutter speed is slow and there is movement within a single frame, not necessarily across the frame. For example, if your shutter speed is at 1/30th of a second and there is someone moving in your scene, that person will come out blurry in the image. So, motion blur is caused when the shutter is left open or “dragged” when objects are moving in the scene, while ghosting is caused when we combine three images with moving objects into a single image.

Image space resolution is inversely proportional to the sensor’s pixel size. This means the smaller the pixel size, higher will be the image space resolution value.

Imagecomics godland

In HDR photography, we are usually taking bracketed sequences to create the final HDR image. Since bracketing involves multiple consecutive shots, any moving objects in your scene will be moving across each bracketed image, a common artifact in HDR photography known as ghosting. Do not confuse ghosting with motion blur as motion blur and ghosting are two different artifacts created by two different pieces of the overall HDR process. In this article, we will explain the difference between motion blur and ghosting in HDR photography.

Ghosting is caused by the combination of the 3 shots with moving objects while motion blur is caused by the moving object with a slow shutter speed. The speed of the moving object and the shutter speed will determine the amount of motion blur and ghosting that will be caused in the image. Although some photographers intentionally use the ghosting artifact to create effects in their images, ghosting may not always be intentional. If necessary, you can use Photoshop or other editing software to fix ghosting across a bracketed sequence.

Ghosting and motion blur are both visible in the image below. However, the motion blur also creates a cool effect with the brake lights, giving the impression that the cars are moving quickly.

In the IMA test method – which is the automated method – the LW/PH value is calculated by considering the MTF30 value. This is an objective measurement of sharpness, which is better than the subjective analysis. The data in the IMA chart is in the units LW/PH, and it can be converted to lp/mm as shown in the above calculation.

As shown in the chart, you can count the number until the line pair degrades. It is where you can’t distinguish the black and white lines (due to merge, grey colors will appear). Typically, the values of this line will be mentioned in the resolution chart (100 x per picture height).

As mentioned before, ghosting is caused when objects moving across multiple images are combined into one image. Ghosting is an actual artifact created from the overall HDR processing. The most common example of ghosting is a bird flying through the scene. In a bracketed sequence, the shots are taken consecutively. Therefore, in the first shot of the bracketed sequence, the bird will be in Position A. In the next shot, the bird will be in Position B and in the last shot, the bird will be in Position C. The bird has moved right across the frame, so when we combine and stack these 3 images, we basically get duplicates of the bird. Each of these duplicates will look a little faded and transparent, making the duplicates of the bird look like little ghosts, hence the name “ghosting.” You will essentially end up with 3 birds in your final HDR image, so you will have to go back and fix each one of the images in the bracketed sequence. With ghosting, there is nothing you can really do, except to shoot a single-shot HDR whenever possible. If you can capture and compress everything to one single image, then shoot a single-shot HDR. If you must do a bracketed sequence, you will need to mask and clone out any ghosting in the HDR image in Photoshop. However, one factor that can help reduce ghosting in HDR photography is to use a DSLR with a faster frame rate, which we will discuss in a future article.

As you can see in the image below, there is a blur where the cars are moving. In addition, the brake lights on the cars are being dragged out because of the slow shutter speed. When you have motion blur in your HDR image, you will always have ghosting as well. We will discuss why this is in the next section of the article.

Vinoth Rajagopalan is an embedded vision expert with 15+ years of experience in product engineering management, R&D, and technical consultations. He has been responsible for many success stories in e-con Systems – from pre-sales and product conceptualization to launch and support. Having started his career as a software engineer, he currently leads a world-class team to handle major product development initiatives

Calculating the resolving power value of a camera lens practically involves taking into consideration the real aspect ratio of the camera.

Ghosted imagemeaning

To calculate the resolving power or object space resolution value using the human eye perception method, you need to first find the LW/PH (line width per picture height) value. To do this, you must observe the line pair highlighted in red (horizontal or vertical) in the resolution chart given in the previous section (Figure 1).