Denso had 26 patents in artificial intelligence during Q2 2024. Denso Corp filed patents in Q2 2024 for methods and devices related to lidar data processing, information code reading, occupant information in mobile objects, and radar system calibration. The lidar data processing method involves using CNN routines to identify objects based on bounding boxes and confidence scores. The information code reading method adjusts learning model parameters based on training data and successfully reads information from images. The device for mobile objects specifies occupant information and estimates requests based on the occupants’ information records. The radar system calibration method involves defining operational characteristics, generating range-azimuth maps, and adjusting sensor characteristics based on intensity values. GlobalData’s report on Denso gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Denso had no grants in artificial intelligence as a theme in Q2 2024.
Recent Patents
Application: Systems and methods for detecting objects based on lidar data (Patent ID: US20240193959A1)
The patent filed by Denso Corp describes a method for detecting objects using lidar data from a vehicle's lidar sensor. The method involves generating lidar inputs, applying a convolutional neural network (CNN) routine to create bounding boxes, determining dimensional characteristics and confidence scores for each box, selecting target boxes based on confidence scores, and identifying objects based on these target boxes. The system includes processors and computer-readable mediums to execute these instructions, with additional features like image-based and point cloud-based lidar inputs, non-maximum suppression (NMS) routines, and Intersection over Union (IoU) routines for accuracy.
The method and system outlined in the patent focus on efficiently detecting objects using lidar data by generating lidar inputs, applying CNN routines to create bounding boxes, and determining dimensional characteristics and confidence scores for each box. The use of NMS and IoU routines enhances the accuracy of object detection by selecting target boxes and evaluating overlaps with ground truth bounding boxes. By combining these techniques, the method and system aim to improve the reliability and effectiveness of object detection in autonomous vehicles based on lidar data.
To know more about GlobalData’s detailed insights on Denso, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
