require 3 characters or more.
About information presented in this cross reference
The information presented in this cross reference is based on TOSHIBA's selection criteria and should be treated as a suggestion only. Please carefully review the latest versions of all relevant information on the TOSHIBA products, including without limitation data sheets and validate all operating parameters of the TOSHIBA products to ensure that the suggested TOSHIBA products are truly compatible with your design and application.
Please note that this cross reference is based on TOSHIBA's estimate of compatibility with other manufacturers' products, based on other manufacturers' published data, at the time the data was collected.
TOSHIBA is not responsible for any incorrect or incomplete information. Information is subject to change at any time without notice.
Select Product Categories
Select Application
Find everything you need for your next product design. Simply select an application and click through to the block diagram to discover our semiconductor solutions.
New Products / News
Pedestrian detection at night-time and 3D reconstruction using a monocular camera: TMPV7608XBG
The TMPV7608XBG is new addition to the Visconti™4 (TMPV760) Series optimized for use in next-generation advanced driver assistance systems (ADAS), which require new vehicles to have a safety feature designed to avoid pedestrian collisions in both day time and night time conditions that will be part of the European New Car Assessment Programme (Euro NCAP).
The TMPV7608XBG allows an ADAS to realize nighttime pedestrian detection as reliable as daytime pedestrian detection available with conventional vision systems. Toshiba’s original Enhanced CoHOG (Co-occurrence Histograms of Oriented Gradients) accelerators combine luminance-based CoHOG feature descriptors with color-based feature descriptors obtained using a newly developed technique. This enhancement leads to a remarkable improvement in the recognition accuracy, especially at nighttime and at scenes with less luminance differences between objects and the background. Additionally, the Enhanced CoHOG accelerators deliver outstanding computational performance, reducing the time taken for recognition.
The TMPV7608XBG incorporates a Structure from Motion (SfM) accelerator that allows detection of general stationary obstacles such as fallen objects and landslides. The SfM accelerator provides three-dimensional (3D) estimates of the distance to, and the height and width of, the stationary obstacles, based on a sequence of images from a monocular camera. This accelerator makes it possible to detect any stationary obstacles without a learning curve, as well as moving objects (using motion analysis) and a particular class of objects such as pedestrians and vehicles (using pattern recognition). The use of 3D information helps reduce the region of interest (ROI) in recognized objects and thus improve the pattern recognition speed, delivering faster driving assistance and control.
The TMPV7608XBG incorporates eight new image processing engines called Media Processing Engines (MPEs) with a double-precision floating-point unit, and various image processing accelerators, numbering 14 in all, to achieve 10 times* the image processing performance of the conventional product. Consequently, the TMPV7608XBG allows parallel execution of up to eight image recognition applications. All the image processing engines and accelerators can recognize pedestrians and vehicles simultaneously in 50 milliseconds, running at a clock frequency of up to 266.7 MHz.
Series Name / Product Name |
|
|
---|---|---|
CPUs | Toshiba original 32-bit RISC Media embedded Processor (MeP) x 2 sets |
|
Image Processing Engines | Toshiba original 32-bit RISC Media Processing Engine Media Processing Engine (MPE) with Double-precision FPU x 8 sets with 256 KB L2 cache x 2 sets |
|
Image Processing Accelerators | Affine Transformation | 3 ch (Lens Undistortion, Affine Transformation, Viewpoint Conversion, etc.) |
Filter | 2 ch (Smoothing, Edge Detection, Corner Detection, Oriented Gradient Computation, Color Space Conversion, Demosaicing, etc.) | |
Histogram | 1 ch (Histogram, Image Conversion using LUT, etc.) | |
Histogram of Oriented Gradients(HOG) | 1 ch (HOG/CoHOG Features + Linear Support Vector Machine) | |
Enhanced CoHOG | 2 ch (HOG/CoHOG/Color-CoHOG Features etc. + Linear Support Vector Machine) | |
Matching | 2 ch (Tracking, Motion Detection) | |
Pyramid | 2ch (Pyramid Image Generation) | |
Structure from Motion (SfM) | 1ch (3D Reconstruction) | |
On-chip Memories | Masked ROM | 80 KBytes |
SRAM | 1632 KBytes | |
On-chip Peripherals | Video Input Interface | Up to 8 ch (CMOS I/F: Up to 4 ch / max. SXGA, 60fps | MIPI® CSI-2 I/F: Up to 4ch / max. 1080p, 60fps) with Video Switch (8-input to 4-select) |
Video Output Interface | 1 ch | |
External Memory Controllers | DDR SDRAM Controller LPDDR2-800 | |
Serial NOR Memory Controller (Single or Quad SPI) | ||
Parallel SRAM / NOR Memory Controller | ||
External MCU Interface | 1 ch (8-bit bidirectional parallel input/output) | |
General-Purpose I/O | 24-bits | |
CAN Controllers | 3 ch | |
I2C Interface | Up to 8 ch | |
UART Interface | Up to 5 ch (1ch for debug use only) | |
SPI Interface | Up to 4 ch | |
PCM Interface | Up to 2 ch | |
Timer / Counters | 32 ch | |
Operating Frequency | 266.7 MHz max. | |
Operating Voltage | Core: 1.1 V, I/O: 3.3 V, 1.2 V (for LPDDR2 SDRAM) / 1.2V (MIPI® CSI-2 I/F) |
|
Package | P-FBGA 796-ball, 27mm x 27mm, 0.8 mm ball pitch |
View the product list of image recognition processors to find products that meet your performance specifications.
Stock Check & Purchase
require 3 characters or more.