Capturing an event at the right moment is what makes an image so valuable. Whether it's a photo of a car running a red light or a snapshot of an item to be inspected on an assembly line, timing is everything. In order to ensure that an event is captured at the appropriate time (within microseconds), a camera’s ability to deterministically trigger is vital. Various imaging solutions will operate with differing triggering reliability, repeatability, and delay, so it is important to choose a camera that will fulfill your timing requirements.
Cameras are instructed to capture an image through the use of an input trigger. This trigger usually comes from one of two sources. The first is a hardware trigger, often in the form of an electrical impulse or signal. The second comes from the computer or system driving the camera and is known as a software trigger. Both of these triggers will have the same end result; however their implementation can be quite different.
When a software trigger is sent, the message can traverse a combination of networks, drivers, buses, and the camera's CPU to be interpreted as a signal to trigger the camera's shutter. System implementation, network congestion, the robustness of the drivers, and the camera's overall processing speeds are all factors that impact the amount of time it takes the signal to reach the camera and tell it to trigger. These elements all introduce a degree of variance to the trigger delay time and contribute to the margin of timing error between the trigger and start of exposure.
Hardware triggers have the advantage of bypassing the network and infrastructure supporting the camera and connect directly to the camera, usually via a purpose-built interface. This results in a much shorter trigger delay and faster camera response. By sending an electrical signal instead of a software-encoded message, layered transmission delays between camera and host can be bypassed, significantly reducing the trigger delay variance and ensuring deterministic triggering.
In timing-crucial applications, deterministic triggering is required to greatly increase the efficacy of the solution. This is why solution designers tend to stray from point and shoot and DSLRs as they can have long software trigger delays with a large variance. These cameras are built for consumer level imaging and little engineering effort is dedicated to building robust drivers, SDKs and APIs to handle software triggers. Moreover, these software based tools can change from one camera model to the next, causing integration issues when newer models are released.
Some DSLR camera models have a software trigger delay of up to 250 ms. If they are used in a red light enforcement system, and a car runs a red light travelling at 80 km/h (50 mph), the car would have travelled 5.5 meters (18 ft) before an image is taken. This is one of the many reasons that industrial solution designers turn to industrial camera manufacturers with technical expertise to produce application-specific imaging tools.
Lumenera has a dedicated team of engineers who strive to create a complete imaging solution for industrial applications. Camera drivers, SDK and API undergo incremental changes and never break legacy functionality to ensure ease of integration and maintenance. Through Lumenera's API, one can trigger some of our cameras with a minimal delay of 6 ms. If even faster triggering is required, Lumenera cameras have hardware triggers that have a delay of as little as 32 µs. It is through advantages like these that industrial grade cameras enhance the performance and reliability of an imaging solution.
