Looking to the future of lidar sensors through the lens of CCTV's past


In recent years, the advancement of lidar technology has significantly transformed real-time data acquisition and processing. As the industry evolves, one pressing question arises: How can we accelerate the lidar’s adoption by making sensors smarter and reducing complexity?

It is a question worth posing in our field, especially with the current technological landscape rapidly evolving. As this evolution has unfolded, lidar technology has become increasingly important to our social development. Consider the usage of lidar applications in autonomous vehicles and urban planning, digitizing and automating the way we work and live. More and more industries are adopting lidar for real-time data acquisition and processing. Hence, attempting to understand the trajectory of sensor technology’s evolution is imperative and necessary.

Our core mission and daily exercise is developing a more efficient, responsive, and adaptable lidar solution for 3D sensors. As a software development company, failing to failing to understand our role in exploring and understanding the potential for intelligent sensor in this evolving market risks stagnation and hinders our ultimate goal.

Without running the risk of idle speculation when postulating possible futures, we can reference the market trends of a technology we are all familiar with that holds some parallels to lidar: CCTV.

Both technologies were considered complex and required specialized knowledge to install and operate when they were first conceived; they have since undergone varying degrees of democratization and standardization. CCTV faced great challenges managing the deluge of data that came with this democratization, as the amount of video data generated increased dramatically; lidar, by nature, is even more demanding in the processing of its data. 

So, with these similarities in mind, let’s have a look at a quick timeline of the CCTV market post-2000:

  • 2000s to Mid-2010s: The transformation from analogue to digital recording systems marked a significant leap forward for CCTV technology. This period saw the replacement of VCRs (Video Cassette Recorders) with digital recorders, drastically improving the ease of data storage and retrieval. CCTV becomes more user-friendly and widely adopted across various sectors​​.

  • 2002: The introduction of CCTV security standards, such as BS8418 in the UK, helped regulate and guide the industry towards consistent quality and reliability​​.

  • 2014: IP cameras, which transmit footage via the Internet, began to overtake analogue systems. This shift was crucial as it allowed digital cameras and devices to integrate seamlessly with network systems, marking a digital transformation in CCTV technology​​.

  • 2021: Leaders of the industry introduce advanced hardware and analytics to their systems, leveraging complex compression techniques and sophisticated software to improve efficiency and performance

  • 2022-2023: The rise of AI-powered analytics contributes majorly to the market, with Tenda’s innovative motion-identification and tracking solutions breaking new ground and Hanwha releasing the world’s first true serverless camera system.

Tracing CCTV’s compelling journey from standalone, analogue systems towards more digitised, interconnected, plug-and-play, AI-driven solutions makes the trajectory of the future of lidar sensors clearer.

This trajectory is now accelerated with lidar’s newfound accessibility. No longer is it an esoteric, obscure and expensive tech only used for a few niche research projects. Modern lidar sensors now come with popular software integrations, through software providers like us here at Flasheye. Enabling a broad range of users to explore new opportunities for their operations, solving many problems in any unique environment.

However, while the parallels between CCTV and lidar allow us to make a reasonable estimate of potential growth, we would be remiss not to mention the unique challenges lidar systems face compared to CCTV systems. The most significant roadblock is the sheer volume of data lidar sensors must handle; it must process a whole other degree of dimensionality than CCTV. This is not to say that CCTV footage analysis is not complex, but tasks such as motion detection and facial recognition require less computational overhead than analysing three-dimensional point clouds, especially at higher resolutions.

This challenge makes the role of data compression methods critical. They are essential for reducing bandwidth and storage requirements, which streamline the transmission and analysis of sensor data. Although CCTV systems are computationally less complex, we can still glean insights from the significant strides they have made in data compression for their datasets.

Additionally, there have been hybrid approaches where data is pre-processed on the sensor to reduce volume and sent to an external processor (like a PC) for further analysis. This pre-processing can include tasks like filtering, compression, and preliminary analysis, helping to mitigate the data bottleneck.

In summary, managing lidar’s data deluge poses a formidable task and is the key to unlocking the potential of smart sensors. Given the sophisticated processing and management strategies already employed, innovation in lidar sensors must continue. We see how this approach has benefited the CCTV market. By doing so for lidar, we unlock new potentials in applying this technology, ultimately contributing to safer, more efficient, and more intelligent systems for our professional and social benefit. Paired with Flasheye’s ability to provide an open software platform and the key building blocks for any sensor on the market, we can push lidar forward together.

Jordan Shepherd - 24/01/2023