AI techniques for blinking light detection for vehicle applications
AssigneeQualcomm Auto Ltd.
InventorAndreas Sjoedin Haellstrand
Life of this patent
Estimating the term…
Abstract
This disclosure provides systems, methods, and devices for vehicle driving assistance systems that support image processing. In a first aspect, a method of image processing includes receiving a first image frame and determining a portion of the image frame that contains a candidate blinking light source. A model may be used to determine a first encoding based on the portion of the first image frame. A second model may use a a sequence of encodings to determine a lighting determination for the first candidate blinking light source. The sequence of encodings may include the first encoding. Other aspects and features are also claimed and described.
Claims30
A method for image processing for use in a vehicle assistance system, comprising:
- receiving a first image frame;
- determining a subset of the first image frame that contains a first candidate blinking light source;
- determining, with a first machine learning model, a first encoding representing visual features within the subset of the first image frame;
- and
- determining, with a second machine learning model and based on a sequence of encodings, a lighting determination for the first candidate blinking light source, wherein the sequence of encodings includes the first encoding,
- wherein the first machine learning model is selected, based on the first candidate blinking light source, from among a plurality of machine learning models, the plurality of machine learning models including a third machine learning model corresponding to a first type of candidate blinking light source and a fourth machine learning model corresponding to a second, different, type of candidate blinking light source.
An apparatus, comprising:
- a memory storing processor-readable code;
- and
- at least one processor coupled to the memory, the at least one processor configured to execute the processor-readable code to cause the at least one processor to perform operations including:
- receiving a first image frame;
- determining a subset of the first image frame that contains a first candidate blinking light source;
- determining, with a first machine learning model, a first encoding representing visual features within the subset of the first image frame;
- and
- determining, with a second machine learning model and based on a sequence of encodings, a lighting determination for the first candidate blinking light source, wherein the sequence of encodings includes the first encoding,
- wherein the first machine learning model is selected, based on the first candidate blinking light source, from among a plurality of machine learning models, the plurality of machine learning models including a third machine learning model corresponding to a first type of candidate blinking light source and a fourth machine learning model corresponding to a second, different, type of candidate blinking light source.
A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform operations comprising:
- receiving a first image frame;
- determining a subset of the first image frame that contains a first candidate blinking light source;
- determining, with a first machine learning model, a first encoding representing visual features within the subset of the first image frame;
- and
- determining, with a second machine learning model and based on a sequence of encodings, a lighting determination for the first candidate blinking light source, wherein the sequence of encodings includes the first encoding,
- wherein the first machine learning model is selected, based on the first candidate blinking light source, from among a plurality of machine learning models, the plurality of machine learning models including a third machine learning model corresponding to a first type of candidate blinking light source and a fourth machine learning model corresponding to a second, different, type of candidate blinking light source.
A vehicle comprising:
- an imaging system configured to capture images from the vehicle;
- and
- a detection system configured to:
- receive a first image frame from the imaging system;
- determine a subset of the first image frame that contains a first candidate blinking light source;
- determine, with a first machine learning model, a first encoding representing visual features within the subset of the first image frame;
- and
- determine, with a second machine learning model and based on a sequence of encodings, a lighting determination for the first candidate blinking light source, wherein the sequence of encodings includes the first encoding,
- wherein the first machine learning model is selected, based on the first candidate blinking light source, from among a plurality of machine learning models, the plurality of machine learning models including a third machine learning model corresponding to a first type of candidate blinking light source and a fourth machine learning model corresponding to a second, different, type of candidate blinking light source.
Dependencies are inferred from the claim text.