INDEX

Machine Vision and the Impact of Artificial Intelligence

For many years, machine vision was extremely limited in its application due to required highlight-controlled environments, expensive sensor technology, and restrictive feature detection technology. The advent of new Deep Learning (DL)-based machine vision technology is set to change the dynamics of this technology and ignite the industry, creating new classes of applications and enormous market opportunities. Machine vision technology is in the process of transition and dramatic expansion.

In this report, ABI Research takes a deeper look at the technology developments that have taken place in the machine vision space, the technical challenges in implementing machine vision, the new applications enabled by DL-based AI, and the total size of the machine vision market opportunity. This report gives a holistic overview and assessment of vendors in every segment of the machine vision technology stack: sensor manufacturers, chipset vendors, software suppliers, DL-Frameworks, and the end market applications and purchasers of the technology.

The report covers significant vendors in the space, which includes machine vision sensor vendors Cognex, Basler AG, Siemens, Keyence, Velodyne, and OMRON.  Companies also covered range from the established and new player supplying chipsets, Nvidia, Intel, Quick Logic, Xilinx, Graphcore, Cambricon, ARM, CEVA, NXP, and Qualcomm, to machine vision software vendors AWS, Clarifai, Ever AI, SenseTime, Megavi Face ++, Irida Labs, MVTec, Deep Vision, Wrnch, and Cognex.

 

Table of Contents

  • 1. EXECUTIVE SUMMARY
  • 2. INTRODUCTION
    • 2.1. Background
  • 3. THE TECHNOLOGY STACK
  • 4. SENSOR TECHNOLOGY
    • 4.1. 2D Cameras
    • 4.2. 3D Cameras
    • 4.3. Other Sensors
    • 4.4. Sensor Vendors List
    • 4.5. Impact of CNNs on Sensors
  • 5. MACHINE VISION SOFTWARE ALGORITHMS
    • 5.1. Early Machine Vision Techniques
    • 5.2. The Next Wave of Machine Vision
    • 5.3. Moving towards 3D Understanding, Not Just 2D Recognition
    • 5.4. Open Source Image Recognition Training Datasets
  • 6. FRAMEWORKS AND MACHINE VISION
    • 6.1. Benchmarking Frameworks for Machine Vision
    • 6.2. Trends
  • 7. HARDWARE PROCESSING VENDORS
    • 7.1. NVIDIA
    • 7.2. Intel
    • 7.3. Arm
    • 7.4. CEVA
    • 7.5. QuickLogic
    • 7.6. Xilinx
    • 7.7. Graphcore
    • 7.8. Hailo
    • 7.9. SambaNova Systems
    • 7.10. Groq
    • 7.11. Horizon Robotics
    • 7.12. Cambricon
    • 7.13. AWS DeepLens
  • 8. SOFTWARE PLATFORM VENDORS
    • 8.1. Machine Vision Software Vendors List
    • 8.2. Amazon Rekognition
    • 8.3. Google Cloud Vision
    • 8.4. Clarifai
    • 8.5. Ever AI
    • 8.6. SenseTime
    • 8.7. Megvii Face++
    • 8.8. IRIDA Labs
    • 8.9. Deep Vision
    • 8.10. MVTec
    • 8.11. wrnch
  • 9. MACHINE VISION USE CASES
    • 9.1. Automotive
    • 9.2. Healthcare
    • 9.3. Retail
    • 9.4. Consumer
    • 9.5. Surveillance
    • 9.6. Robotics
    • 9.7. Industrial
    • 9.8. Commercial
    • 9.9. Insurance
  • 10. MACHINE VISION TRENDS AND FORECASTS
    • 10.1. Research Methodology
    • 10.2. Overview
    • 10.3. Automotive
    • 10.4. Industrial
    • 10.5. Retail
    • 10.6. Surveillance
    • 10.7. Smartphones
    • 10.8. Robotics
    • 10.9. Smart Home
    • 10.10. Investment
  • 11. RECOMMENDATIONS
    • 11.1. Precedent
    • 11.2. Privacy
    • 11.3. Data Quality
    • 11.4. Cost
  • 12. CONCLUSIONS