IMAGE PROCESSOR - VISION PROCESSOR - INDUSTRY OVERVIEW

Image and vision for AI, a unusual ecosystem

Extracted from:
- Image Signal Processor and Vision Processor Market and Technology Trends report and Artificial Intelligence Computing for Automotive report, Yole Développement, 2019
- ZF S-Cam 4 – Forward Automotive Mono and Tri Camera for Advanced Driver Assistance Systems and Mobileye EyeQ4 Vision Processor Family report, System Plus Consulting, 2019

OUTLINES:

  • AI-powered newcomers are reshuffling the pack.
  • Visualization: algorithms needed to transform the raw data into a visible image by the human eye have existed for a long time and are optimized in terms of performance or quality.
  • Analysis: new algorithms require computing power to achieve the precision sought in understanding the surrounding environment...
  • AI has completely disrupted hardware in vision systems, and has had an impact on entire segments, as Mobileye has in automotive, for example.
  • ZF, one of the largest tier one suppliers of automotive systems, last year released its fourth Generation S-Cam with two solutions, one with a mono camera and the other with a triple camera set-up.


TO DOWNLOAD THE PRESS: 
ENGLISH

LYON, France – September 10, 2019: “AI has completely disrupted hardware in vision systems, and has had an impact on entire segments,” comments Yohann Tschudi, PhD. Technology & Market Analyst at Yole Développement (Yole).
A good example, is the penetration of Mobileye within the automotive market segment. Yole and its partner System Plus Consulting deeply analyzed a scenario for AI within the dynamics of the autonomous automotive market, especially with Mobileye’s market positioning, and present an understanding of AI’s impact on the semiconductor industry with two dedicated reports: Artificial Intelligence Computing for Automotive and Mobileye EyeQ4 Vision Processor Family.
Today, without doubts, image analysis adds a lot of value. Image sensor builders are therefore increasingly interested in integrating a software layer to their system in order to capture it.
“Image sensors must go beyond taking images – they must be able to analyze them”, adds Yohann Tschudi from Yole.
The market research and strategy consulting company, Yole pursues its investigation all year long, in the computing and software world. Analysts propose today a new technology and market report, titled Image Signal Processor and Vision Processor Market and Technology Trends. This new report, part of the software & computing reports collection focuses on describing the markets related to hardware needed for image processing. Behind a camera, there may be several ways to process raw data depending on the purpose. The alternatives usually break down into viewing or analyzing the image to understand the environment around the module or system containing the camera. Each of these purposes, however, requires a different type of hardware. Under this report, Yole’s analysts segment processing and computing respectively according to their association with the image signal processor and vision processor. At the business level, segmentation is quite simple. Some companies offer a license and royalties for a design, which is known as IP business. Other companies sell the chips, which we call the silicon business. Image signal processor, vision processor, deep learning, computing… What are we talking about? Yole and System Plus Consulting invite you to discover an overview of this industry.

To run dedicated software, high power computing and memory are necessary, which led to the creation and development of vision processors. The ISP market offers a steady 3% CAGR from 2018 to 2024, making the total market worth US$4.2 billion in 2024. Meanwhile, the vision processor market is exploding, with a 18% CAGR between 2018 and 2024, making the market worth US$14.5 billion in 2024.

Beyond market figures, this is a completely new industry that has been created, especially with AI-powered newcomers. These companies are today reshuffling the pack of this industry.

Processing and computing hardware for the imaging market has been divided into two different business models. IP companies do not have physical products, but silicon companies sell the physical processors. The leaders are easy to identify for each category. ARM and Synopsys lead the IP segment and OmniVision, Mobileye and ON Semiconductor lead the silicon segment.
Yole’s team had the opportunity to debate with Andy Hanvey, Director of Marketing, Automotive at OmniVision about the latest market trends and technical issues. During this discussion, he detailed OmniVision’s vision for the future of image processing and its strategy. Discover the full interview on i-Micronews.com.
The company is in the forefront of the scene thanks to its portfolio of image sensors in automotive sector and related ISPs. As an example, System Plus Consulting, sister company of Yole, did a detailed teardown of ZF’s fourth generation ADAS S-Cam. This camera is using OmniVision’s OV10642 image sensor in combination with the latest Mobileye EyeQ4 vision processor.
“ZF, one of the largest tier one suppliers of automotive systems, last year released its fourth Generation S-Cam with two solutions, one with a mono camera and the other with a triple camera set-up,” explains Audrey Lahrach, Cost Analyst at System Plus Consulting. “These cameras feature the Omnivision CMOS image sensor, which demonstrates the shift in the procurement strategies of ZF and Intel Mobileye. In fact, the use of the latest Mobileye EyeQ4 vision processor allows new sensors to be used and makes the S-Cam4 one of the smallest and lightest products in its category…”

With this latest software & computing analysis, Yole’s Semiconductor & Software team developed an impressive expertise in this domain. Therefore, this report is an opportunity to understand what is happening with the emergence of AI.
“Even if it is not a new technology, thanks to technological factors AI has made a spectacular entry into vision systems,” asserts Yohann Tschudi. “It opens new perspectives in mobile device, automotive, computing and surveillance industries.” Applications include biometry and photography, autonomous driving, behavioral recognition, human identification and tracking.

So what about this dedicated ecosystem? Indeed it is important to note that historical players have struggled to react to AI’s arrival. That has allowed other companies to get into the business, including smartphone companies like Apple and Huawei, startups like Mobileye, and companies in other segments, like NVIDIA in automotive applications. However, because the trend is towards low-power, low-consumption, always-on computing hardware, the historical players are coming back into the game. AI technologies promise a bright future in many areas, with rapid software and hardware progress. This is very exciting for the entire area of vision systems.

A detailed description of the software & computing collection of reports is now available on i-Micronews.com, software & computing reports collection.

Acronyms:
AI : Artificial Intelligence
IP : Intellectual Property
ISP : Image Signal Processor
CAGR : Compound Annual Growth Rate