JP-WO2025164369-A5 -

Dates

Publication Date

20260511

Application Date

20250117

Description

To achieve the above objective, the first image recognition device of this disclosure comprises an image processing unit (103) capable of multitasking, which performs multiple image recognition tasks on an image, and which can adjust the processing content of the image recognition tasks, and a controller unit (104, 104a, 104b) that adjusts the processing content of the multiple image recognition tasks in the image processing unit. The controller unit takes an image as input and dynamically changes the processing content of the multiple image recognition tasks in the image processing unit according to the trends of the image content . The controller unit has a scene classification unit (1041) that classifies the scene indicated by the image content, which takes an image as input, and a first uncertainty prediction unit (1042) that predicts the uncertainty of the scene classified by the scene classification unit. The controller unit dynamically changes the processing content of the multiple image recognition tasks in the image processing unit according to the scene classified by the scene classification unit as a trend of the image content, and also dynamically changes the processing content of the multiple image recognition tasks in the image processing unit using the uncertainty of the scene predicted by the first uncertainty prediction unit . To achieve the above objective, the second image recognition device of this disclosure is usable in a vehicle and comprises an image processing unit (103) capable of multitasking, which performs multiple image recognition tasks on an image and adjusts the processing content of the image recognition tasks, and a controller unit (104, 104a, 104b) which adjusts the processing content of the multiple image recognition tasks in the image processing unit. The controller unit takes an image as input and dynamically changes the processing content of the multiple image recognition tasks in the image processing unit according to the trends in the content of the image. It also comprises a vehicle-related information acquisition unit that acquires vehicle-related information other than images, and the vehicle-related information acquisition unit acquires at least vehicle speed information. The controller unit takes an image as input and dynamically changes the processing content of the multiple image recognition tasks in the image processing unit according to the trends in the content of the image as well as the vehicle speed information acquired by the vehicle-related information acquisition unit. To achieve the above objective, the third image recognition device of this disclosure is usable in a vehicle and comprises an image processing unit (103) capable of multitasking, which performs multiple image recognition tasks on an image and can adjust the processing content of the image recognition tasks, and a controller unit (104, 104a, 104b) which adjusts the processing content of the multiple image recognition tasks in the image processing unit, wherein the controller unit takes an image as input and dynamically changes the processing content of the multiple image recognition tasks in the image processing unit according to the trends in the content of the image, and comprises a vehicle-related information acquisition unit that acquires vehicle-related information other than images, the vehicle-related information acquisition unit acquires at least information on the driver's state of the vehicle, and the controller unit takes an image as input and dynamically changes the processing content of the multiple image recognition tasks in the image processing unit according to the trends in the content of the image as well as the driver's state information acquired by the vehicle-related information acquisition unit. To achieve the above objective, the first image recognition method of this disclosure includes an image processing step that is capable of multitasking, executing multiple image recognition tasks on an image, and is performed by at least one processor, and is capable of adjusting the processing content of the image recognition tasks, and a controller step that adjusts the processing content of the multiple image recognition tasks in the image processing step, wherein the controller step takes an image as input and dynamically changes the processing content of the multiple image recognition tasks in the image processing step according to the trends of the image content, wherein the controller step includes a scene classification step that classifies the scene indicated by the image content, and a first uncertainty prediction step that predicts the uncertainty of the scene classified in the scene classification step, wherein the processing content of the multiple image recognition tasks in the image processing step is dynamically changed according to the scene classified in the scene classification step as a trend of the image content, and the processing content of the multip