Many people overlook “triggers.”

The production line isn’t very fast, and the cameras are also limited to 60 fps or 120 fps.

Logically speaking, it should be sufficient.

But it still occurs on site:

One item fewer, the position is not fixed, and images are occasionally missing.

At this point, the issue may not be the frame rate, but rather:

Trigger method.

In industrial vision applications, when a product passes a designated location, the camera typically needs to capture an image at the precise moment.

Shooting continuously doesn’t necessarily mean you’ll capture the “correct position.”

If the camera continuously captures images:

30fps, 60fps, 120fps……

The camera just keeps taking pictures at its own pace.

But when exactly does the product happen to come into view of the camera,

The camera doesn’t know.

So it may occur:

The previous product hasn’t arrived yet,

The next one is already over.

That’s why many pipeline projects use:

Trigger photo capture.

What is a trigger?

It can be simply understood as:

The product has arrived; I’ll take another photo.

The most common practice is:

The product passes through a photoelectric sensor,

The sensor emits a signal,

The camera takes a photo immediately upon receiving the signal.

Typical workflow: The sensor detects the product → Sends a trigger signal → The light source turns on → The camera captures an image.

What is the difference between software triggering and hardware triggering?

Software trigger

Computer sends a command:

“Let’s take a photo now.”

Simple configuration, suitable for:

Static testing, debugging, and projects with low time constraints.

However, computer systems and program communication themselves introduce a certain degree of latency and variability.

Hardware Trigger

Sensors, PLCs, and encoders directly supply electrical signals to the camera.

As soon as the signal arrives, the camera begins capturing.

When software triggering lacks sufficient precision or when hardware signals are already present on-site, hardware triggering can be used.

Therefore, in high-speed assembly lines, what is more common is:

Photoelectric sensor / PLC / Encoder + hardware trigger

Why are encoders still used on high-speed production lines?

If the conveyor belt speed varies, relying solely on a fixed-time delay may not be sufficiently stable.

At this point, you can use:

Encoder.

The encoder moves along with the conveyor belt,

The camera determines when to take a shot based on the actual distance of movement.

Sensors detect the products, encoders determine how far the products have moved, and cameras capture images at the appropriate positions.

This method is particularly suitable for:

High-speed conveying, positioning and imaging, sorting,Line-scan inspection.

Another common reason for missed shots is triggering too quickly.

Assumption:

The previous image is still being exposed,

The next trigger signal has arrived again.

The camera may not have time to respond.

Basler’s documentation explicitly states that if the previous frame’s exposure has not yet completed and a new exposure trigger is issued, the new trigger may be ignored, resulting in an overtrigger.

So:

The trigger rate must also not exceed the camera’s actual processing speed.

This is also why:

120An FPS camera ≠ it’s fine to trigger 120 times per second at random.

It must also take into account the exposure time, readout time, and the actual configuration.

Finally, remember this one sentence.

The frame rate determines how fast the camera can shoot.

It triggers the decision of when the camera takes a shot.

On the industrial assembly line,

Often, “when will it be filmed,”

It’s more important than “how many shots per second it can take.”

If the following occurs on site:

Missing shots, positional drift, and inconsistent product positioning across images.

Don’t just focus on FPS.

First, check:

Trigger signal, sensor position, exposure time, trigger frequency.

Many so-called “camera misses,”

Actually:

The trigger isn’t configured properly.