Dehua Song

Title: Dehua Song: Innovator in Neural Network Technology

Introduction

Dehua Song is a prominent inventor based in Moscow, Russia. He has made significant contributions to the field of neural networks and image processing. His innovative work has led to the development of a unique patent that enhances the capabilities of neural networks.

Latest Patents

Dehua Song holds a patent for a "Neural network training method, image processing method, and apparatus." This patent outlines a method that includes obtaining an input feature map of a training image. It involves performing feature extraction processing using a feature extraction core of a neural network to obtain a first candidate feature map. The method further includes adding the first candidate feature map and a second candidate feature map to generate an output feature map. The second candidate feature map is derived by increasing the value corresponding to each element in the input feature map by N times, where N is greater than 0. The process determines an image processing result based on the output feature map and adjusts the neural network's parameters accordingly.

Career Highlights

Dehua Song is currently employed at Huawei Technologies Co., Limited, where he continues to push the boundaries of technology. His work at Huawei has allowed him to collaborate with other talented individuals in the field, contributing to advancements in neural network applications.

Collaborations

Some of his notable coworkers include Hanting Chen and Yunhe Wang. Their collaborative efforts have further enriched the research and development environment at Huawei.

Conclusion

Dehua Song's innovative contributions to neural network technology exemplify the impact of dedicated inventors in the tech industry. His patent reflects a significant advancement in image processing methods, showcasing the potential of neural networks in various applications.