Systems and methods for detecting objects in images
AssigneeZhejiang Dahua Technology Co., Ltd.
InventorsXin Ye, Songlin Yang
Life of this patent
Estimating the term…
Abstract
A method configured to implemented on at least one image processing device for detecting objects in images includes obtaining an image including an object. The method also includes generating one or more feature vectors related to the image based on a first convolutional neural network, wherein the one or more feature vectors includes a plurality of parameters. The method further includes determining the position of the object based on at least one of the plurality of parameters. The method still further includes determining a category associated with the object based on at least one the plurality of parameters.
Claims19
An image processing system configured to detect objects in images, comprising:
- a bus;
- at least one storage medium connected to the bus, including a set of instructions for object detection;
- and
- logic circuits in communication with the at least one storage medium via the bus, wherein when executing the set of instructions, the logic circuits:
- obtain first electronic signals including an image including an object;
- generate one or more feature vectors related to the image based on a first convolutional neural network, the one or more feature vectors including a plurality of parameters;
- determine a position of the object based on at least one of the plurality of parameters;
- determine a category associated with the object based on at least one of the plurality of parameters;
- determine that the object includes a predetermined scene;
- and
- obtain a first feature of the object in the image based on a second convolutional neural network, wherein the predetermined scene includes a person close to a window of a vehicle;
- and the first feature includes whether the person is fastened with a seat belt.
A method configured to implemented on at least one image processing device to detect objects in images, comprising:
- obtaining, by the at least one image processing device, an image including an object;
- generating, by the at least one image processing device, one or more feature vectors related to the image based on a first convolutional neural network, wherein the one or more feature vectors includes a plurality of parameters;
- determining, by the at least one image processing device, a position of the object based on at least one of the plurality of parameters;
- determining, by the at least one image processing device, a category associated with the object based on at least one the plurality of parameters;
- determining that the object includes a predetermined scene;
- and
- obtaining a first feature of the object in the image based on a second convolutional neural network, wherein the predetermined scene includes a person close to a window of a vehicle;
- and the first feature includes whether the person is fastened with a seat belt.
A non-transitory medium storing instructions, the instructions, when executed by at least one image processing device, causing the at least one image processing device to implement a method, comprising:
- obtaining, by the at least one image processing device, an image including an object;
- generating, by the at least one image processing device, one or more feature vectors related to the image based on a first convolutional neural network, wherein the one or more feature vectors includes a plurality of parameters;
- determining, by the at least one image processing device, a position of the object based on at least one of the plurality of parameters;
- and
- determining, by the at least one image processing device, a category associated with the object based on at least one the plurality of parameters;
- determining that the object includes a predetermined scene;
- and
- obtaining a first feature of the object in the image based on a second convolutional neural network, wherein the predetermined scene includes a person close to a window of a vehicle;
- and the first feature includes whether the person is fastened with a seat belt.
Dependencies are inferred from the claim text.