Asset Details
Do you wish to reserve the book?
High-Performance Detection of Concealed Forbidden Objects on Human Body with Deep Neural Networks Based on Passive Millimeter Wave and Visible Imagery
Hey, we have placed the reservation for you!
By the way, why not check out events that you can attend while you pick your title.
Oops! Something went wrong.
Looks like we were not able to place the reservation. Kindly try again later.
Are you sure you want to remove the book from the shelf?
High-Performance Detection of Concealed Forbidden Objects on Human Body with Deep Neural Networks Based on Passive Millimeter Wave and Visible Imagery
Do you wish to request the book?
High-Performance Detection of Concealed Forbidden Objects on Human Body with Deep Neural Networks Based on Passive Millimeter Wave and Visible Imagery
Please be aware that the book you have requested cannot be checked out. If you would like to checkout this book, you can reserve another copy
We have requested the book for you!
Your request is successful and it will be processed during the Library working hours. Please check the status of your request in My Requests.
Oops! Something went wrong.
Looks like we were not able to place your request. Kindly try again later.
Journal Article
High-Performance Detection of Concealed Forbidden Objects on Human Body with Deep Neural Networks Based on Passive Millimeter Wave and Visible Imagery
2019
Overview
In this paper, a high-performance detection algorithm of concealed forbidden objects on human body is presented based on deep neural networks (DNN) and complementary advantages of passive millimeter wave imagery (PMMWI) and visible imagery (VI). With well capacity of penetrability, PMMWI can effectively reveal suspected forbidden objects concealed on human body without harm of ionizing radiation compared with conventional X-ray methods. However, due to its current limited imaging capability, the resolution of PMMWI is still unsatisfactory and easy to result in false alarms. Therefore, by complementarity of superiorities, VI is employed to overcome the deficiency of confusions. In this way, massive image samples of PMMWI and VI are simultaneously acquired and manually annotated as necessary training datasets to carry out deep learning on DNN models so as to achieve high-performance human body profile segmentation on both PMMWI and VI. Then, high-precision region registration of human body profiles is implemented between PMMWI and VI to localize and confirm high confident suspected targets and remove false alarm regions as well. According to the principle of synthetic integration and global optimization, a high performance detection algorithm system is constructed, analyzed, and assessed. A series of comprehensive experiment results demonstrate the outstanding performance of our proposed detection algorithm.
Publisher
Springer US,Springer Nature B.V
