Member Info for johnchucka

Ford 2019-11-21/ Toyota 2019-11-21 / GM 2019-11-28 patents all describe distortion analysis and rectification.
'Unprecedented Collaboration' or JConfirmation Bias ? you decide.

FORD GLOBAL TECH LLC 2019-11-14
DISTORTION CORRECTION FOR VEHICLE SURROUND VIEW CAMERA PROJECTIONS
Method and apparatus are disclosed for distortion correction for vehicle surround view camera projections. An example vehicle includes cameras to capture images of a perimeter around the vehicle and a processor. The processor, using the images, generates a composite image of an area around the vehicle, and generates a depth map that defines spatial relationships between the vehicle and objects around the vehicle. The processor also generates a projection surface using the depth map. Additionally, the processor presents an interface for generating a view image based on the composite image projected onto the projection surface.
https://worldwide.espacenet.com/publicationDetails/biblio?II=3&ND=3&adjacent=true&locale=en_EP&FT=D&date=20191114&CC=US&NR=2019349571A1&KC=A1#

S7, how do you imagine Continentals Conti/Xilinx attitude to the Bosch offer. Conti have also increased their engineering capacity over the last 18 months and Xilinx have committed further to SEE so perhaps the offer the rejection was a clear fork in the road for SEE/ Conti/ Xilinx etc.

Semicast Research@semicast_res
Great to see CEO Paul McGlone on stage to accept this award in person too. In a world of “move fast and break things”, inclusion in the tech industry really matters.
https://twitter.com/RachelSS_MLA/status/1202694955677863937

More on the Camera calibration theme.
GM
THREE-DIMENSIONAL ALIGNMENT OF RADAR AND CAMERA SENSORS 2019-04-26
As previously noted, vehicles increasingly include sensor systems such as radar sensors and camera sensors. In autonomous vehicles or vehicles with autonomous features (eg, autonomous parking), 360-degree coverage around the vehicle with more than one sensor facilitates the acquisition of supplementary information through sensor fusion. However, sensor fusion (ie, a combination of data acquired by each sensor) requires geometric alignment of the individual sensors that share the FOV. If the sensors are not aligned, the detection of a frame of reference that is converted to other sensors by one sensor will be projected at the wrong coordinates. For example, radar detections that are converted to the camera coordinate system are projected at the wrong image coordinates. Therefore, the distance (in pixels) between the projected position and the actual image position is a measure of the sensor's misalignment.

Embodiments of the systems and methods described in detail herein relate to 3D alignment of radar sensors and camera sensors. Specifically, the conversion parameters between the radar and the camera are determined to align the two types of sensors geometrically. Then, the radar detection converted to the camera reference frame is projected onto the target image at the correct image coordinates. In the exemplary embodiment described in detail herein, corner reflectors are used to determine the conversion parameters. In radar systems, corner reflectors appear as strong point targets, where all reflected energy comes from near the apex. By inserting a light emitting diode (LED) in the vertex of the corner reflector, the image coordinates of the LED in the image acquired by the camera can be aligned with the vertex detection performed by the radar system, as described in detail above.

According to an exemplary embodiment, FIG. 1 is a block diagram of a vehicle 100 that performs 3D alignment of a radar sensor 110 and a camera sensor 120. The exemplary vehicle 100 shown in FIG. 1 is a car 101. In addition to the radar 110 and the camera 120, the vehicle 100 may include additional sensors (eg, lidar, additional radar, additional cameras), where the FOV is guided to cover the sides and rear of the vehicle 100. The vehicle 100 includes a controller 130 for aligning the radar 110 and the camera 120. For example, the controller 130 may be a stand-alone controller or may be part of one of the sensors or part of the automation system of the vehicle 100. The controller 130 includes a processing circuit, which may include an application specific integrated circuit (ASIC), an electronic circuit, a processor (shared, dedicated or grouped) and memory executing one or more software or firmware programs, a combination logic circuit, and / Or other suitable components that provide the described functionality.
https://tinyurl.com/wrbb3oa

https://seeingmachines.springboard.com.au/jobtools/jncustomsearch.searchResults?in_organid=18900&in_jobDate=All

It's all about Calibration

Method for operating a driver assistance system for motor vehicles 2019-12-05
The invention relates to a method for operating a driver assistance system for motor vehicles, which has a radar sensor and has a variable range of functions, with continuous recalibration of the radar sensor for compensation of adjustment errors during operation.
The function of a driver assistance system for motor vehicles is frequently based on the location data of a radar sensor installed in the front of the vehicle, with which distances, relative speeds, azimuth angles and possibly also elevation angles of vehicles ahead and other objects in the surroundings of the vehicle are measured. Based on these data, the driver assistance system then provides various assistance functions, such as automatic proximity control, collision warning functions, automatic initiation of emergency braking for collision avoidance, and the like. An error-free and safe operation of the driver assistance system requires an accurate assessment of the traffic environment, including correct assumptions about the azimuth angles of the located objects. Correct angle measurements are possible with the help of the radar sensor, however, only if the sensor has been correctly adjusted when installed in the vehicle, so that the direction corresponding to the azimuth angle measurement of 0 ° is really the forward direction of the vehicle.
https://worldwide.espacenet.com/publicationDetails/biblio?CC=US&NR=2019369234A1&KC=A1&FT=D&ND=3&date=20191205&DB=&locale=en_EP#

TELEMATICS DEVICE AND METHOD FOR TAKING OUT VEHICLE INSURANCE AND ESTIMATING INSURANCE FEES 2019-12-05
A method operating a telematics device of a vehicle can include: requesting identity (ID) information of a driver from a driver terminal when the driver enters the vehicle; performing an authentication process using the ID information; preventing start-up of an engine of the vehicle when the driver is identified as a temporary user according to the authentication process; receiving navigation data including information relating to a destination from the driver terminal; acquiring information indicating an estimated insurance fee corresponding to each of one or more paths for navigating to the destination; transmitting the information indicating the estimated insurance fee to the driver terminal; and starting the engine when one of the one or more paths for navigating to the destination is selected by the driver terminal.

According to embodiments of the present disclosure, a method of operating a telematics device of a vehicle can include: requesting identity (ID) information of a driver from a driver terminal in communication with the telematics device when the driver enters the vehicle; performing an authentication process using the ID information; controlling operation of the vehicle such that start-up of an engine of the vehicle is prevented when the driver is identified as a temporary user according to the authentication process; receiving navigation data including information relating to a destination from the driver terminal; acquiring information indicating an estimated insurance fee corresponding to each of one or more paths for navigating to the destination, wherein the estimated insurance fee is calculated based on the navigation data, external environment data, and driver data; transmitting the information indicating the estimated insurance fee to the driver terminal; and controlling operation of the vehicle such that the engine is started when one of the one or more paths for navigating to the destination is selected by the driver terminal.

The requesting of the ID information may include detecting whether a door of the vehicle is closed, a hand of the driver touches a steering wheel of the vehicle, or a start button of the vehicle is pressed; and requesting the ID information upon detecting that the door of the vehicle is closed, the hand of the driver touches the steering wheel of the vehicle, or the start button of the vehicle is pressed.
https://patents.justia.com/patent/20190366980

We have seen an upswing in SEE employing Data analysts / Labelling roles etc . Perhaps related

SYSTEM AND METHOD FOR GENERATING IMPROVED SYNTHETIC IMAGES 2019-12-05
System, methods, and other embodiments described herein relate to improving the generation of realistic images. In one embodiment, a method includes acquiring a synthetic image including identified labels of simulated components within the synthetic image. The synthetic image is a simulated visualization and the identified labels distinguish between the components within the synthetic image. The method includes computing, from the simulated components, translated components that visually approximate real instances of the simulated components by using a generative module comprised of neural networks that are configured to separately generate the translated components. The method includes blending the translated components together to produce a new image from the simulated components of the synthetic image.

As an additional note, while the adaptation system 100 is generally applicable to many different contexts within which machine learning algorithms may be implemented, in one or more embodiments, the adaptation system 100 is implemented to at least support functions of an advanced driving assistance system (ADAS) and/or an autonomous driving module that provides autonomous driving (i.e., self-driving) functionality to a vehicle. For example, the autonomous driving module, in one embodiment, is comprised of a plurality of sub-modules that each perform a respective task in relation to autonomously operating the vehicle. That is, the autonomous driving module includes sub-modules that, in combination, determine travel path(s), current autonomous driving maneuvers for the vehicle 600, future autonomous driving maneuvers and/or modifications to current autonomous driving maneuvers. In general, to effectively operate the individual task-specific modules need to be trained using labeled data from different situations that the vehicle may encounter.

However, as previously noted, acquiring this training data, especially when targeted to a particular task module, can represent a significant difficulty. Therefore, the adaptation system 100 can be employed to generate training data for training various sub-modules by translating synthetic images with task-specific subject matter into photo-realistic images. Accordingly, the adaptation system 100, in one embodiment, provides training data (e.g., labeled photorealistic images) to one or more of the sub-modules or components thereof as manner of more efficiently generating training data.
https://patents.justia.com/patent/20190370666

For reference. No explicit reference to in/out cameras found. Posted to demonstrate Patent evolution with an assumption that SEE patents are potentially related/play a role in the Camera sensor fusion calibration algorithms. This maybe the completely wrong rabbit hole.

INFERRING LOCATIONS OF 3D OBJECTS IN A SPATIAL ENVIRONMENT 2019-12-05
A method for inferring a location of a three-dimensional (3D) object, the method includes receiving sensor data from a plurality of sensors of an autonomous vehicle. The method also includes mapping features extracted from the sensor data to a first data structure corresponding to a feature space of a 3D representation of a spatial area. The method further includes classifying the merged features to identify an object within a vicinity of the autonomous vehicle and projecting a 3D representation of the identified object to a location of the 3D feature space. The method still further includes controlling an action of the autonomous vehicle based on the projected 3D representation.

Background
Autonomous agents (e.g., vehicles, robots, drones, etc.) and semi-autonomous agents use machine vision for analyzing areas of interest in a surrounding environment. Although scientists have spent decades studying the human visual system, a solution for realizing equivalent machine vision remains elusive. Machine vision is distinct from the field of digital image processing due to the desire to recover a three-dimensional (3D) structure of the world from images and using the 3D structure for fully understanding a scene. That is, machine vision strives to provide a 3D view of the environment that surrounds an autonomous agent.
In operation, autonomous agents may rely on a trained convolutional neural network (CNN) to identify objects within areas of interest in an image of a surrounding environment. For example, a CNN may be trained to identify and track objects captured by one or more sensors, such as light detection and ranging (LIDAR) sensors, sonar sensors, red-green-blue (RGB) cameras, RGB-depth (RGB-D) cameras, and the like. The sensors may be coupled to, or in communication with, a device, such as the autonomous agent. Object detection applications for autonomous agents may analyze sensor image data for detecting objects (e.g., pedestrians, cyclists, other cars, etc.) in the surrounding scene from the autonomous agent.
https://patents.justia.com/patent/20190371052

Honda 2019-11-14
EXTERNAL-WORLD RECOGNITION SYSTEM
[0001] The present invention relates to an external environment recognition system (external-world recognition system) that acquires external environment information around a vehicle with an external environment sensor, such as a camera, and recognizes the external environment information with an external environment recognition device. More particularly, it relates to an external environment recognition system that corrects the direction in which the external environment recognition device performs recognition.

[0004] The posture of the camera changes due to vibration of the vehicle, contact with an occupant, and the like. Along with a change in the posture of the camera, the optical axis direction of the camera and the reference direction of the external environment recognition device become misaligned with the direction to be recognized. This causes an error in position information of a recognition target determined from a recognition result of the external environment recognition device. Japanese Laid-Open Patent Publication No. 2011-169728 does not take into consideration correcting of a recognition result of the external environment recognition device.

[0005] The present invention was made in view of such a challenge and an object thereof is to provide an external environment recognition system capable of accurately correcting a reference direction established for an external environment recognition device and accurately recognizing the positions of recognition targets around a vehicle.

[0006] An external environment recognition system according to a first aspect of the present invention includes: a satellite positioning device configured to measure a position of a vehicle by receiving radio waves transmitted from artificial satellites; a vehicle sensor configured to detect or estimate a behavior of the vehicle; an external environment sensor configured to acquire external environment information around the vehicle; and an external environment recognition device configured to recognize the external environment information centered in a reference direction, wherein the external environment recognition device is configured to correct a detected value of the vehicle sensor based on positions of the vehicle in temporal sequence measured by the satellite positioning device and correct the reference direction based on the corrected detected value.
https://worldwide.espacenet.com/publicationDetails/description?CC=US&NR=2019347878A1&KC=A1&FT=D&ND=3&date=20191114&DB=&locale=en_EP#

Veoneer at CES 2020: showcasing the products that will change how everyone drives in the future

'Some of the other core technologies that will be highlighted, displayed and demonstrated by Veoneer at CES are next generation vision cameras and radars, 5G technology in partnership with Ericsson and Ericsson Connected Vehicle Cloud to enable and enhance Veoneer solutions, driver and in-cabin monitoring, and next generation thermal imaging'.
https://www.marketscreener.com/VEONEER-INC-44211877/news/Veoneer-at-CES-2020-showcasing-the-products-that-will-change-how-everyone-drives-in-the-future-29684021/

S2020,
A question to add to your question.
Do the 'over the air updates' give any opportunity for IP protection?

LEWBO,
Lets look forward progress on that front to see if the association I made ON 21 Nov 2019 07:39 GM / Hyundai Patent HYUNDAI MOTOR CO LTD [KR]; KIA MOTORS CORP [KR] 2019-08-22 Comparison holds firm.
Note that a later patent continues to develop the innovation of inward /outward camera fusion which also looks to be running in parallel with Honda Patents technologically speaking. Maybe that's why Harish is smiling.

APPARATUS AND METHOD FOR GENERATING TIME-LAPSE IMAGE USING CAMERA IN VEHICLE 2019-06-17
According to the present invention, an apparatus for generating a time-lapse video using a camera in a vehicle comprises: a video acquiring unit configured to acquire an image of surroundings of a vehicle and a video of a driver; a video processing unit configured to generate a time-lapse video by processing the acquired video of surroundings of a vehicle to enable the same to have a first speed; a detection unit configured to detect a speed of the vehicle and a gaze direction of the driver; a navigation configured to measure a location of the vehicle; a control unit configured to control a play rate of the time-lapse video based on the location and speed of the vehicle and the gaze direction of the driver; and a display on which the time-lapse video is outputted. Accordingly, when a gaze of the driver deviates from the front or the speed of the vehicle decreases, it is possible to record details of an outside scenery of the vehicle by lowering the play rate of the time-lapse video. When the gaze of the drive keeps ahead or the vehicle drives at a high speed, it is possible to shorten the video by increasing the play rate of the time-lapse video.
https://worldwide.espacenet.com/publicationDetails/biblio?CC=KR&NR=20190067380A&KC=A&FT=D&ND=3&date=20190617&DB=&locale=en_EP#

Jazz2020,
Interesting remark their at the end of the video by Paul,
'Our engineering management capabilities have been hugely strengthened under Danny's Leadership.'
That would be Danny Edmunds (ex Conti) who was one of the First Xilinx/Conti fusion team and his appointments as we have seen look to have involved bringing more Conti staff over no doubt assisted and guided by Gary Collins (ex Conti) a auto industry grandee.

S7 ..I think working out that S2020 is in fact also Jazz has got to stand out as the high water mark in your research journey.

As Prime Mover Neil Elwood Peart once said 'the point of the journey is not to arrive'.

Good find S2020,
That's what the patents have been describing and my guess based on that is that SEE are now a essential algorithm provider in both inward and outward facing fusion and calibration for all the AVCC players . Of course they have not told us about this small detail yet and it's critical role in developing the next global baseline safety architecture or it's value to the company. No doubt too busy building shelves and choosing dusters for all those shinny awards to deal with these trivial items of news.

'For example, many safety automation features require multiple, forward-looking sensors. Furthermore, these must encompass both radar and camera-based technology, as well as innovative fusion capabilities to combine them. That’s because the two types of sensor offer distinct characteristics. Radar does not need light, and its operation is not compromised by conditions such as fog. Additionally, radar is not able to distinguish between different types of objects in the way that a camera-based sensor can. Traditionally, all these sensors would have a sophisticated built-in CPU to process data, adding to the overall system cost. But as part of Aptiv’s goal to democratize safety, the company has developed a new approach: satellite architecture'

Looking at the time line of Anes Hodžic patent ' System for displaying the environment of a vehicle according to the direction of the driver's look' which has a Priority date of 2016-12-20 and then reading across to a SEE patent CAMERA REGISTRATION IN A MULTI-CAMERA SYSTEM which has a priority date of 2016-06-29 one might then see this period as the early stages of the Inward/outward facing Camera fusion being developed to compliment the sensor systems we are now seeing described in GM/TOYOTA/Hyundai/ Continental/ Honda patents aka the ' accelerated deployment of IoT devices and automotive platforms on a global scale'

'In his final position at Bosch, Anes was responsible for Bosch Connected Services business, driving development and accelerated deployment of IoT devices and automotive platforms on a global scale'.

Dr Anes Hodžic, From Bosch driver Gaze/external camera Patents guy to Airbus Global VP of IoT

System for displaying the environment of a vehicle according to the direction of the driver's look . 2018-06-29
A system for displaying the environment of a vehicle is disclosed and comprises two external cameras for taking pictures and/or videos of the vehicle environment, at least one display device in the passenger compartment of the vehicle to represent the views taken by the external cameras, a control device for controlling the representation on the at least one display device and an internal camera of the passenger compartment to determine the direction of the driver's gaze. The control device controls the representation of the views taken by the external cameras on the at least one display device (11) according to the determined direction of the driver's gaze.
https://worldwide.espacenet.com/publicationDetails/biblio?II=0&ND=3&adjacent=true&locale=en_EP&FT=D&date=20180629&CC=CN&NR=108216034A&KC=A#

Dr Anes Hodžic
Global VP of IoT

As the Head of Internet of Things (IoT) within the Airbus Digital Transformation Office, Anes is responsible for overseeing Airbus’ IoT initiatives. This includes technology roadmap definition, solution industrialisation, customer support, and deployment strategy and execution across all divisions. In this role, Anes focuses on demonstrating IoT’s business value and accelerating its scaling.

Before joining Airbus in 2017, Anes held a variety of executive engineering positions at Bosch Group. There, he worked on innovation management covering a broad technology portfolio in a cross-industrial environment, as well as corporate strategy development targeting new business development. During his tenure at Bosch, Anes helped to introduce innovative control technologies in hybrid and electric cars, developed connectivity solutions, UX design and self-driving (ADAS) solutions for trucks, cars and motorcycles. In his final position at Bosch, Anes was responsible for Bosch Connected Services business, driving development and accelerated deployment of IoT devices and automotive platforms on a global scale.

Anes holds a doctorate degree in electrical engineering and telecommunications from the Technical University in Berlin, Germany, and an executive MBA degree from Ashridge Business School in London, UK.

https://www.industryofthingsworld.com/peoples/dr-anes-hodzic/
https://www.linkedin.com/in/aneshodzic/detail/recent-activity/

Cenkos imo will serve SEE management with which ever dish they want with very little consideration for well researched contradictory evidence from this board or Safestocks. They'll cover their expensive blandness with a nice flowery 'conservative' wall paper. Objectivity for PI's is the last thing they are interested in imo.

Always interested Sch. Thanks