Left: Pixel chip. Right: Logic chip

Sony and Prophesee Develop a Stacked Event-Based Vision Sensor with the Industry’s Smallest Pixels and Highest HDR Performance

Left: Pixel chip. Right: Logic chip

Figure 1: Chip photographs. Left: Pixel chip. Right: Logic chip

SHARE:





February 19, 2020

Tokyo, Japan—Sony Corporation and Prophesee S.A. today announced they have jointly developed a stacked event-based vision sensor with the industry's smallest 4.86μm pixel size and the industry's highest 124dB (or more) HDR performance. The new sensor and its performance results were announced at the International Solid-State Circuits Conference (ISSCC) held in San Francisco in the United States, starting on February 16, 2020.

The new stacked event-based vision sensor detects changes in the luminance of each pixel asynchronously and outputs data including coordinates and time only for the pixels where a change is detected, thereby enabling high efficiency, high speed, low latency data output. This vision sensor achieves high resolution, high speed, and high time resolution despite its small size and low power consumption. This accomplishment was made possible by combining technical features of Sony's stacked CMOS image sensor, resulting in small pixel size and excellent low light performance that are achieved by the use of Cu-Cu connection, with Prophesee's Metavision® event-based vision sensing technologies leading to fast pixel response, high temporal resolution and high throughput data readout. The newly developed sensor is suitable for various machine vision applications, such as detecting fast moving objects in a wide range of environments and conditions.

Main Features

  1. Small size and high resolution delivered by stacked event-based vision sensor with the industry's smallest 4.86μm pixel size

    The pixel chip (top) and the logic chip (bottom) incorporate signal processing circuits which detect changes in luminance based on an asynchronous delta modulation method are arrayed separately. Each pixel of the two individual chips is electrically connected using Cu-Cu connection in a stacked configuration. In addition to the industry's smallest 4.86μm pixel size, the sensor also delivers 1/2 type, 1280x720 HD resolution by achieving high density integration with a fine 40nm logic process.

  2. Industry's highest 124dB (or more) HDR performance achieved by high aperture ratio

    The industry's highest 124dB (or more) HDR performance is made possible by placing only back-illuminated pixels and a part of N-type MOS transistor on the pixel chip (top), thereby allowing the aperture ratio to be enhanced by up to 77%. High sensitivity/low noise technologies Sony has developed over many years of CMOS image sensor development enable event detection in low-light conditions (40mlx).

    Pixel illustration
    Figure 2: Pixel illustration. Left: Conventional Sensor. Right: Developed Sensor

  3. Event data readout with high time resolution and high output

    While a frame-based sensor outputs entire images at fixed intervals according to the frame rate, an event-based sensor selects pixel data asynchronously using a row selection arbiter circuit. By adding time information at 1μs precision to the pixel address where a change in luminance has occurred, event data readout with high time resolution is ensured. Furthermore, a high output event rate of 1.066Geps has been achieved by efficiently compressing the event data, i.e. luminance change polarity, time, and x/y coordinate information for each event.

    Event data readout illustration
    Figure 3: Event data readout illustration.

    Captured image example 1
    Figure 4: Captured image example 1.
    Left: Conventional sensor. Right: Event-based sensor

    * Captured at night

    Captured image example 2
    Figure 5: Captured image example 2
    Left: Conventional Sensor. Right: Event-based sensor

    * Captured at night

About Prophesee

Prophesee develops the world's most advanced neuromorphic vision systems. The company's Event-Based approach to machine vision allows for significant reductions of power, latency and data processing requirements as compared to traditional frame-based vision systems.
Prophesee's sensors and algorithms are designed to mimic the way biological eyes and brains work - with the goal to significantly improve performance, power and data efficiency of artificial vision systems in areas such as autonomous vehicles, industrial automation, IoT, security and surveillance, and AR/VR.

prophesee.ai

Microsoft
Development Category (English)728x90