WO2023052448A1 - Procédé de détection d'obstacles - Google Patents
Procédé de détection d'obstacles Download PDFInfo
- Publication number
- WO2023052448A1 WO2023052448A1 PCT/EP2022/077017 EP2022077017W WO2023052448A1 WO 2023052448 A1 WO2023052448 A1 WO 2023052448A1 EP 2022077017 W EP2022077017 W EP 2022077017W WO 2023052448 A1 WO2023052448 A1 WO 2023052448A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- rectified
- images
- disparity
- obstacle detection
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 30
- 230000011218 segmentation Effects 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims description 33
- 238000001514 detection method Methods 0.000 claims description 30
- 230000003190 augmentative effect Effects 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 230000007246 mechanism Effects 0.000 claims description 7
- 238000012795 verification Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 230000010354 integration Effects 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 230000003287 optical effect Effects 0.000 description 8
- 230000004044 response Effects 0.000 description 6
- 238000012549 training Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 101150061340 Ier2 gene Proteins 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 2
- 230000010006 flight Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 230000033001 locomotion Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 229940050561 matrix product Drugs 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000012995 silicone-based technology Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/17—Terrestrial scenes taken from planes or by drones
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/593—Depth or shape recovery from multiple images from stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/70—Labelling scene content, e.g. deriving syntactic or semantic representations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
- G06T2207/10021—Stereoscopic video; Stereoscopic image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Definitions
- the invention relates to the field of obstacle detection to provide assistance in piloting a vehicle or a robot.
- Such an obstacle detection system has the particular objective of reducing collision accidents as much as possible by providing the driver or the navigation system of the vehicle with a certain amount of information relating to the obstacles present in the environment of the vehicle: position obstacles, distance between each obstacle and the vehicle, size and type of obstacles, etc.
- Obstacles are, for example, electric cables, pylons, trees, etc.
- Obstacle detection systems are thus known that use RADAR technology (for RADio Detection And Ranging) or LIDAR technology (for Light Detection And Ranging).
- PODS Passive Obstacle Detection System
- MWIR bands for Mid-Wave InfraRed
- LWIR bands for Long-Wave InfraRed
- PMMW sensors for Passive MilliMeter Wave
- Still other systems use satellite images, but are not intended to be embedded in vehicles such as a helicopter or a drone.
- the object of the invention is to detect and identify various types of obstacles precisely and reliably, without emitting electromagnetic or light waves, in order to provide assistance in piloting any type of vehicle.
- a method for detecting obstacles is proposed, implemented in at least one processing unit and comprising the steps of:
- raw stereo images representative of a environment of a vehicle and produced by stereo cameras, the raw stereo images comprising a left raw image and a right raw image;
- the obstacle detection method according to the invention detects and differentiates obstacles accurately and reliably.
- the method for detecting obstacles according to the invention can provide assistance in piloting any type of vehicle and does not require the emission of electromagnetic or light waves.
- the augmented images can for example be projected onto the visor of a pilot's helmet.
- a method for detecting obstacles is also proposed as previously described, in which the rectification step comprises a distortion correction and uses first parameters comprising extrinsic and intrinsic parameters of the stereo cameras.
- a method of detecting obstacles as previously described is further proposed, in which epipolar lines of the rectified stereo images and of the rectified segmented stereo images are horizontal.
- the integration step comprises the steps, for each predefined obstacle instance, of determining, using coordinates of the obstacle instance predefined and the three-dimensional coordinates of the reconstructed pixels, a distance between said predefined obstacle instance and the vehicle as well as dimensions of said predefined obstacle instance.
- the integration step comprises, for each instance of predefined obstacle, the step of embedding a cross on a barycenter of said instance of predefined obstacle.
- a method for detecting obstacles as previously described is also proposed, in which the stereo cameras are infrared cameras.
- a system comprising stereo cameras and a processing unit in which the method for detecting obstacles as previously described is implemented.
- a computer program is also proposed comprising instructions which lead the processing unit of the system as previously described to execute the steps of the method for detecting obstacles as previously described.
- FIG. 1 schematically represents different modules of a helicopter in which the invention is implemented;
- FIG. 2 represents steps of the method for detecting obstacles according to the invention;
- FIG. 3 represents the implementation of the semantic segmentation algorithm
- FIG. 4 represents a U-Net neural network
- FIG. 5 represents an HRNet neural network
- FIG. 6 represents stereo images and makes it possible to illustrate the verification of the disparity
- FIG. 7 represents an object and the optical centers of two cameras
- FIG. 8 illustrates the way in which the 3D coordinates of a pixel are reconstructed
- FIG. 9 represents an image corresponding to a real result of the implementation of the method for detecting obstacles according to the invention.
- a non-limiting embodiment of the invention is described here, in which the method for detecting obstacles according to the invention is used to provide assistance to the pilot of a helicopter.
- the helicopter 1 firstly comprises a capture device 2 intended to capture images of the environment of the helicopter 1.
- the capture device 2 here comprises a plurality of infrared stereo cameras 3, in this case two stereo front cameras and two stereo side cameras.
- the cameras 3 comprise left cameras 3a, located on the left - port side - of the helicopter 1 and right cameras 3b, located on the right - starboard side - of the helicopter 1.
- the optical sensors of the cameras 3 are wide-field sensors (for example of 180° class, with 80° of field of view per individual camera).
- the optical sensors are for example of the thermo-detector type using micro-bolometer technology, or of the photodetector type using silicon-based technology.
- the cameras 3 operate here in the far infrared range (LWIR: Long Wave InfraRed).
- LWIR Long Wave InfraRed
- the method for detecting obstacles according to the invention can therefore be used and effective day and night.
- the helicopter 1 also comprises an inertial unit 4 integrating an inertial measurement unit 5 (UNI) and making it possible to estimate the orientation, position and speed of the helicopter 1.
- the UNI 5 integrates navigation sensors comprising at least three accelerometers and three gyrometers or gyroscopes.
- the helicopter 1 further comprises a processing unit 6 comprising at least one processing component 7 which is adapted to execute instructions of a program to implement the obstacle detection method according to the invention.
- the program is stored in a memory 8 of the processing unit 6.
- the processing component 7 is for example a conventional processor, a graphics processor (or GPU, for Graphics Processor Unit), a microcontroller, a DSP (for Digital Signal Processor r which can be translated as “digital signal processor”), or else a programmable logic circuit such as an FPGA (for Field Programmable Gate Arrays') or an ASIC (for Application Specific Integrated Circuit).
- the method of detecting obstacles operates as follows.
- the processing unit 6 acquires raw stereo images Isb in real time.
- the raw stereo images Isb are representative of the environment of the helicopter 1 and are produced by the cameras 3 of the capture device 2.
- the raw stereo images Isb comprise a raw left image and a raw right image.
- the processing unit 6 then implements a semantic segmentation algorithm of the deep learning type, which uses a neural network suitable for image segmentation.
- the implementation of the semantic segmentation algorithm makes it possible to produce a first segmented image lel from a first raw image Ibl, the first raw image Ibl being the left raw image or l raw image on the right.
- the first raw image Ibl is the left raw image.
- the neural network is for example a U-Net network, an HRNet network (for High-Resolution Network), or an HRNet+OCR network (OCR for Object-Contextual Representation).
- the U-Net network 10 is a convolutional neural network, which comprises an encoder part 11 and a decoder part 12.
- the network 10 comprises a network block which processes the response of the image at increasingly smaller successive scales (encoder part), then a network block which processes the response of the image at increasingly larger successive scales (decoder part), before carrying out a semantic segmentation.
- the HRNet network 14 is characterized by the fact of processing the image at different scales at the same time.
- the HRNet + OCR network is the sum of two networks.
- the OCR network is characterized by the use of an attention mechanism which consists in measuring correlations between the channels and the pixels constituting the last blocks of the network.
- the semantic segmentation algorithm was previously trained using a previously established database.
- the weights associated with the neurons are calculated from the database during a training process.
- the database is an annotated database; we also speak for such learning of supervised learning on labeled data.
- the obstacles have been annotated by hand.
- This database includes both images from real flights (acquisition flights) and masks.
- the images come from sequences taken over several hundred hours of flight.
- the images are infrared images.
- Each pixel of the images in the database is encoded to designate a particular class among a plurality of classes.
- the annotated classes were selected to represent two types of object: objects forming obstacles and objects constituting the environment.
- the objects constituting the environment are intended to be identified in order to facilitate the discrimination of obstacles in different environments.
- the discrimination of the environment also makes it possible to identify in the image potential landing zones and the horizon line.
- the plurality of classes here comprises seven classes, which designate respectively the sky, the water, the ground, the vegetation, the cables, the pylons, and the other types of obstacles on the human surface (other than the cables and the pylons) .
- Other classes can be defined, for example a class designating humans and another class designating vehicles.
- the definition of the classes is adapted to the application in which the invention is implemented.
- the classes chosen here make it possible to differentiate types of obstacles which are specifically dangerous for a helicopter, such as cables, and also make it possible to detect obstacles which may constitute passage points, such as pylons.
- a pylon is an indicator of the flight altitude to be maintained to avoid any collision with cables.
- the database was created in such a way as to avoid the redundancy of visual information in the different classes. Certain objects are therefore related to other objects of a different type but having a similar visual appearance.
- a lamp post is not an electric pylon but its visual appearance is such that it can easily be related to it.
- only part of the base is used for learning, while another part is used to verify the results obtained and thus avoid “overfitting” of the network.
- the database has three parts: a “training” part which is directly used in the backpropagation of the gradients to iteratively elaborate the weights of the neurons, a “validation” part which is used to monitor the performance of the neural network on non-standard data. training views, and allows to adjust the training hyper-parameter, and finally a "test” part, which is known only to the client entity / organization.
- the test part can be used by the client entity to measure performance with its own data.
- This inference block is a matrix of dimension CxHxW, where C is the number of classes, H is the height of the input image and W is the width of the input image.
- This inference block is the response of the input image to the neural network.
- Each pixel is then normalized using a Softmax type function. This reflects the response of the image into a matrix where the sum of the C elements of the dimension corresponding to the classes and associated with each particular pixel is equal to approximately 1. This is only a normalized answer and not a probability. This normalized value makes it possible to have visibility on the preponderance of a class compared to the others, simply from its value taken in isolation.
- the coordinate of the element whose response to the system is the highest is extracted from the line corresponding to the classes and associated with said pixel.
- a class is thus associated with each pixel of the first raw image.
- the processing unit 6 then rectifies the raw stereo images Isb to obtain rectified stereo images Isr (pair of rectified images), and the first segmented image lel to obtain a first rectified segmented image lerl.
- the rectification includes distortion correction and uses first PI parameters including extrinsic and intrinsic parameters of the cameras 3.
- the extrinsic and intrinsic parameters were produced during calibration operations carried out in the factory.
- the extrinsic parameters include levels of roll, pitch and yaw of the cameras 3 with respect to the “system” reference (reference associated with the helicopter 1).
- the intrinsic parameters include the focal length of each camera 3, the distance between the cameras 3, distortion parameters, the size of each pixel, the binocular center distance.
- the rectification makes it possible to obtain the rectified stereo images Isr and the first rectified segmented image lerl, whose epipolar lines are horizontal.
- the processing unit 6 then implements an algorithm for calculating the disparity between the rectified stereo images Isr, to obtain a disparity map Cd.
- the disparity is a measurement used to estimate the depth and therefore to reconstruct a view of the environment of the helicopter 1 in three dimensions.
- the disparity map Cd is a digital image (disparity image) which contains information on the correspondences of the points of the same scene taken with two different angles of view, in this case with the cameras on the left 3a and with the right cameras 3b of the capture device 2.
- the images of disparities change in real time.
- the processing unit 6 then checks the disparity value of each pair of homologous pixels comprising a left pixel of a left image (produced by the left cameras 3a of the capture device 2) and a right pixel of a right image (produced by the right cameras 3b of the capture device 2).
- disparity checking eliminates false matches caused by occlusions.
- the pair of homologous pixels comprises a left pixel 15a and a right pixel 15b, which respectively belong to a left image 16a and to a right image 16b (and which have the same vertical y coordinate).
- the coordinate of the right pixel 15b is:
- Xd xg - dispg ( xg ,y), dispg being the disparity estimated by taking the left image 16a as a reference.
- the coordinate of the left pixel 15a is:
- Xg x d +dispd(x d ,y), dispd being the disparity estimated by taking the right image 16b as a reference.
- a mechanism for post-accumulation of the disparity images could be used to verify the disparity and, in particular, a mechanism for post-accumulation of the disparity images.
- This mechanism consists in projecting past disparity images onto the current disparity image, in order to temporally filter the disparity value of each pixel. To do this, we use the disparity values, navigation information from inertial units and GPS information, intrinsic parameters and extrinsic parameters of the cameras. Such a post-accumulation makes it possible to reduce the errors of the disparity.
- Several verification mechanisms could be used in combination.
- the detection method then implements a spatial transformation of the first rectified segmented image Ier1, using the disparity map, to produce a second rectified segmented image Ier2.
- the second rectified segmented image Ier2 corresponds to the side opposite that of the first raw image (that is to say here to the right side). Iser-rectified segmented stereo images are thus produced.
- the spatial transformation here is a spatial interpolation.
- the disparity map Cd is therefore used to transpose the first rectified segmented image lerl (here left side image) onto the eye on the other side (here right eye), by spatial interpolation of lerl with the motion field induced by the disparity, to produce a second rectified segmented image ieler2 (here, image of the right side) and thus constitute a segmented stereo pair Iser from a single segmentation inference on the left eye (computational gain).
- the Isr rectified stereo images and the Iser rectified segmented stereo images are then projected into a frame associated with the helmet of the pilot of helicopter 1.
- the processing unit 6 then implements a three-dimensional reconstruction algorithm, using the disparity map Cd, to produce three-dimensional coordinates for each pixel of the raw stereo images Isb (original images)
- the three-dimensional reconstruction algorithm uses second parameters P2 comprising the extrinsic and intrinsic parameters of the cameras 3, as well as navigation data produced by the navigation sensors of the UMI 5.
- the navigation data comprises roll, pitch and yaw levels of the helicopter 1 in a local geographical reference, as well as roll, pitch and yaw levels of the cameras 3 with respect to the helicopter 1, as well as only one latitude and one longitude of the helicopter 1.
- Three-dimensional reconstruction uses the following principles.
- C-C' is the baseline B (base line) and corresponds to the distance between the optical centers of the cameras C and C'.
- f focal length
- Tx being the baseline
- f being the focal length
- the processing unit 6 thus produces three-dimensional coordinates of pixels of a reconstructed three-dimensional image representative of the environment of the helicopter 1.
- the three-dimensional coordinates of the pixels are defined in a local geographical reference associated with the helicopter 1 and corrected in yaw (thanks to the navigation data).
- the three-dimensional coordinates of the pixels are then projected to produce two-dimensional I2D images of the three-dimensional coordinates of each reconstructed pixel.
- the processing unit 6 analyzes the Iser rectified segmented stereo images and selects the class of each pixel according to its response to the semantic segmentation algorithm.
- the processing unit 6 then performs an instantiation of each obstacle and produces a list of predefined obstacle instances Obst present in the environment of the helicopter 1 from the rectified segmented stereo images Iser.
- the processing unit 6 determines that a group of pixels of a rectified segmented stereo image Iser forms an instance of predefined obstacle Obst when said pixels are connected to each other. It is considered here that two pixels are connected together if one of the two pixels belongs to the neighborhood of the other and if the two pixels belong to the same class.
- the neighborhood of a particular pixel is defined herein as comprising eight "primary" neighboring pixels around said particular pixel.
- the processing unit 6 allocates a specific identifier for each group of pixels which are adjacent to each other in order to be able to classify each obstacle.
- the processing unit 6 then produces a stream comprising the images labeled Ilab as well as the coordinates of all the instances of predefined obstacles Obst.
- the processing unit 6 then integrates, using the three-dimensional coordinates, the instances of predefined obstacles Obst in intermediate images obtained from the rectified stereo images Isr, to produce augmented images Ia intended to provide piloting assistance of helicopter 1.
- the intermediate images are here the Isr rectified stereo images themselves, that is to say that the instances of predefined obstacles Obst are integrated into the Isr rectified stereo images to produce the augmented images la.
- the processing unit 6 first of all implements a calculation algorithm to extract global characteristics on the size of each instance of obstacle predefined in height and width by measuring the standard deviation of the 3D positions of the pixels making up said predefined obstacle instance.
- the calculation algorithm thus acquires the coordinates of each instance of predefined obstacle Obst and uses the two-dimensional I2D images of the three-dimensional coordinates of the reconstructed pixels to calculate, for each instance of predefined obstacle Obst, the average distance between said instance of predefined obstacle and the helicopter, as well as the width and height of said instance predefined obstacle.
- the processing unit 6 indeed knows the pixels belonging to the same obstacle thanks to the segmentation, the 3D reconstruction of each pixel having a valid disparity, and can therefore deduce the characteristics therefrom in the local geographical reference centered on the helicopter 1 and yaw corrected.
- Measuring the physical characteristics of obstacles allows the flight path to be adjusted based on the data.
- the processing unit 6 implements a merging algorithm to merge the instances of predefined obstacles Obst into the rectified stereo images Isr.
- the fusion is carried out by the embedding of a cross on a barycenter of said instance of predefined obstacle, with additional information on its geometric characteristics.
- the pixels belonging to a predefined obstacle instance are highlighted by locally increasing the intensity of each pixel of the instance, or else by inserting a false color. This false color can moreover correspond to a type of obstacle.
- the processing unit 6 thus produces the augmented images 1a (in this case stereo images).
- the augmented images 1a are then projected onto the visor of the pilot's flight helmet.
- the headset is a stereoscopic headset.
- the augmented images are projected in both eyes of the helmet so as to obtain a stereoscopic effect.
- the processing unit 6 carries out an extraction of the pinhole projection (mark where there is no deformation of the 3D world), corresponding to the field of vision of the pilot and to the orientation of his helmet.
- the segmentation algorithm is implemented on a single raw image (left or right), called “first raw image”, corresponding to a single eye.
- the disparity map is then used to project, via a spatial transformation, the segmentation into the unsegmented eye.
- the spatial transformation step is not carried out and the Iser-rectified segmented stereo images are obtained directly at the output of the rectification step.
- the invention is not necessarily implemented to provide assistance in piloting a helicopter, but can be used in any type of vehicle, with or without a pilot: aircraft, land vehicle, drone, any type of robot (se moving on the ground, in the air), etc. Augmented images can be used to achieve autonomous navigation.
- the intermediate images, in which the instances of predefined obstacles are integrated, are not necessarily the Isr rectified stereo images, but could be other images, and for example three-dimensional images.
- the method for detecting obstacles according to the invention is not necessarily implemented in a single processing unit integrated into a vehicle, but can be implemented in one or more processing units, among which one at least less can be located at a distance from the vehicle (and for example in a base from which the vehicle departed, in another vehicle, in a Cloud server, etc.).
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Remote Sensing (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
Description
Claims
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CA3233479A CA3233479A1 (fr) | 2021-09-28 | 2022-09-28 | Procede de detection d'obstacles |
IL311708A IL311708A (en) | 2021-09-28 | 2022-09-28 | A method for detecting obstacles |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FRFR2110221 | 2021-09-28 | ||
FR2110221A FR3127611B1 (fr) | 2021-09-28 | 2021-09-28 | Procédé de détection d’obstacles |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2023052448A1 true WO2023052448A1 (fr) | 2023-04-06 |
Family
ID=80449018
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2022/077017 WO2023052448A1 (fr) | 2021-09-28 | 2022-09-28 | Procédé de détection d'obstacles |
Country Status (4)
Country | Link |
---|---|
CA (1) | CA3233479A1 (fr) |
FR (1) | FR3127611B1 (fr) |
IL (1) | IL311708A (fr) |
WO (1) | WO2023052448A1 (fr) |
-
2021
- 2021-09-28 FR FR2110221A patent/FR3127611B1/fr active Active
-
2022
- 2022-09-28 CA CA3233479A patent/CA3233479A1/fr active Pending
- 2022-09-28 IL IL311708A patent/IL311708A/en unknown
- 2022-09-28 WO PCT/EP2022/077017 patent/WO2023052448A1/fr active Application Filing
Non-Patent Citations (3)
Title |
---|
FUA PASCAL ED - MOESLUND THOMAS B ET AL: "A parallel stereo algorithm that produces dense depth maps and preserves image features", MACHINE VISION AND APPLICATIONS, SPRINGER VERLAG, DE, vol. 6, 1 January 1993 (1993-01-01), pages 35 - 49, XP002475536, ISSN: 0932-8092, DOI: 10.1007/BF01212430 * |
HUH ET AL: "A stereo vision-based obstacle detection system in vehicles", OPTICS AND LASERS IN ENGINEERING, ELSEVIER, AMSTERDAM, NL, vol. 46, no. 2, 13 November 2007 (2007-11-13), pages 168 - 178, XP022342816, ISSN: 0143-8166, DOI: 10.1016/J.OPTLASENG.2007.08.002 * |
RAHUL Y ET AL: "Camera-Based Object Detection, Identification and Distance Estimation", 2018 2ND INTERNATIONAL CONFERENCE ON MICRO-ELECTRONICS AND TELECOMMUNICATION ENGINEERING (ICMETE), IEEE, 20 September 2018 (2018-09-20), pages 203 - 205, XP033567344, DOI: 10.1109/ICMETE.2018.00052 * |
Also Published As
Publication number | Publication date |
---|---|
FR3127611A1 (fr) | 2023-03-31 |
FR3127611B1 (fr) | 2023-11-10 |
IL311708A (en) | 2024-05-01 |
CA3233479A1 (fr) | 2023-04-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110926474B (zh) | 卫星/视觉/激光组合的城市峡谷环境uav定位导航方法 | |
WO2019230339A1 (fr) | Dispositif d'identification d'objet, système de déplacement d'un corps, procédé d'identification d'objet, procédé d'apprentissage d'un modèle d'identification d'objet et dispositif d'apprentissage de modèle d'identification d'objet | |
US11570416B2 (en) | Multichannel, multi-polarization imaging for improved perception | |
JP6574611B2 (ja) | 立体画像に基づいて距離情報を求めるためのセンサシステム | |
KR102200299B1 (ko) | 3d-vr 멀티센서 시스템 기반의 도로 시설물 관리 솔루션을 구현하는 시스템 및 그 방법 | |
JP2022512290A (ja) | 色覚困難の知覚を改善する多重スペクトル多重偏光(msmp)フィルタ処理 | |
CA2851750A1 (fr) | Procede d'etalonnage d'un systeme de vision par ordinateur embarque sur un mobile | |
CN111209840B (zh) | 一种基于多传感器数据融合的3d目标检测方法 | |
CN112561996A (zh) | 一种自主水下机器人回收对接中目标检测方法 | |
Zhao et al. | Environmental perception and sensor data fusion for unmanned ground vehicle | |
US20220383585A1 (en) | Systems And Methods For Generating And/Or Using 3-Dimensional Information With Camera Arrays | |
CN116403191A (zh) | 一种基于单目视觉的三维车辆跟踪方法、装置和电子设备 | |
WO2023052448A1 (fr) | Procédé de détection d'obstacles | |
CN113706599B (zh) | 一种基于伪标签融合的双目深度估计方法 | |
EP1570433B1 (fr) | Dispositif de traitement d'images a reconnaissance et selection de sources de lumiere | |
Nguyen et al. | Neural network‐based optical flow versus traditional optical flow techniques with thermal aerial imaging in real‐world settings | |
GB2582419A (en) | Improvements in and relating to range-finding | |
FR3101467A1 (fr) | Systeme de traitement d'images par mise en correspondance multifonctions et modulaire pour aeronef, et procede associe | |
FR3094081A1 (fr) | Procédé d’estimation passive du temps avant collision pour un aéronef ou de tout objet volant pouvant être guidé, procédé de navigation associé | |
EP4086819A1 (fr) | Procédé d'apprentissage d'une intelligence artificielle supervisée destinée à identifier un objet prédeterminé dans l'environnement d'un aéronef | |
Scaramuzza et al. | Application challenges from a bird's eye view | |
WO2023161568A1 (fr) | Procédé de calcul de surfaces tridimensionnelles pour véhicule équipé d'un système d'aide à la conduite | |
FR3103291A1 (fr) | Méthode de traitement de données reçues de capteurs configurés pour être embaqués dans un véhicule | |
WO2021190910A1 (fr) | Dispositif et procédé d'aide à la décision d'un système cognitif artificiel | |
WO2023111189A1 (fr) | Procédé de détection d'un objet dans l'environnement d'un aéronef, système et programme d'ordinateur associés |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22797069 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 311708 Country of ref document: IL |
|
WWE | Wipo information: entry into national phase |
Ref document number: 3233479 Country of ref document: CA |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2022797069 Country of ref document: EP Effective date: 20240429 |