The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Sebahagian besar sumber pengiraan sistem terbenam untuk pengesanan visual dikhususkan untuk pengekstrakan ciri, dan ini menjejaskan ketepatan pengesanan dan prestasi pemprosesan sistem dengan teruk. Untuk menyelesaikan masalah ini, kami mencadangkan deskriptor ciri berdasarkan histogram kecerunan berorientasikan (HOG) yang terdiri daripada algebra linear mudah yang boleh mengekstrak maklumat setara dengan deskriptor ciri HOG konvensional pada kos pengiraan yang rendah. Dalam penilaian, algoritma pengesanan terdepan dengan vektor terurai HOG (DV-HOG) ini mencapai ketepatan pengesanan yang setara atau lebih baik berbanding dengan deskriptor ciri HOG konvensional. Pelaksanaan perkakasan DV-HOG menduduki kira-kira 14.2 kali lebih kecil kawasan sel daripada pelaksanaan HOG konvensional.
Koichi MITSUNARI
Osaka University
Yoshinori TAKEUCHI
Kindai University
Masaharu IMAI
Osaka University
Jaehoon YU
Osaka University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Koichi MITSUNARI, Yoshinori TAKEUCHI, Masaharu IMAI, Jaehoon YU, "Decomposed Vector Histograms of Oriented Gradients for Efficient Hardware Implementation" in IEICE TRANSACTIONS on Fundamentals,
vol. E101-A, no. 11, pp. 1766-1775, November 2018, doi: 10.1587/transfun.E101.A.1766.
Abstract: A significant portion of computational resources of embedded systems for visual detection is dedicated to feature extraction, and this severely affects the detection accuracy and processing performance of the system. To solve this problem, we propose a feature descriptor based on histograms of oriented gradients (HOG) consisting of simple linear algebra that can extract equivalent information to the conventional HOG feature descriptor at a low computational cost. In an evaluation, a leading-edge detection algorithm with this decomposed vector HOG (DV-HOG) achieved equivalent or better detection accuracy compared with conventional HOG feature descriptors. A hardware implementation of DV-HOG occupies approximately 14.2 times smaller cell area than that of a conventional HOG implementation.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E101.A.1766/_p
Salinan
@ARTICLE{e101-a_11_1766,
author={Koichi MITSUNARI, Yoshinori TAKEUCHI, Masaharu IMAI, Jaehoon YU, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Decomposed Vector Histograms of Oriented Gradients for Efficient Hardware Implementation},
year={2018},
volume={E101-A},
number={11},
pages={1766-1775},
abstract={A significant portion of computational resources of embedded systems for visual detection is dedicated to feature extraction, and this severely affects the detection accuracy and processing performance of the system. To solve this problem, we propose a feature descriptor based on histograms of oriented gradients (HOG) consisting of simple linear algebra that can extract equivalent information to the conventional HOG feature descriptor at a low computational cost. In an evaluation, a leading-edge detection algorithm with this decomposed vector HOG (DV-HOG) achieved equivalent or better detection accuracy compared with conventional HOG feature descriptors. A hardware implementation of DV-HOG occupies approximately 14.2 times smaller cell area than that of a conventional HOG implementation.},
keywords={},
doi={10.1587/transfun.E101.A.1766},
ISSN={1745-1337},
month={November},}
Salinan
TY - JOUR
TI - Decomposed Vector Histograms of Oriented Gradients for Efficient Hardware Implementation
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1766
EP - 1775
AU - Koichi MITSUNARI
AU - Yoshinori TAKEUCHI
AU - Masaharu IMAI
AU - Jaehoon YU
PY - 2018
DO - 10.1587/transfun.E101.A.1766
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E101-A
IS - 11
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - November 2018
AB - A significant portion of computational resources of embedded systems for visual detection is dedicated to feature extraction, and this severely affects the detection accuracy and processing performance of the system. To solve this problem, we propose a feature descriptor based on histograms of oriented gradients (HOG) consisting of simple linear algebra that can extract equivalent information to the conventional HOG feature descriptor at a low computational cost. In an evaluation, a leading-edge detection algorithm with this decomposed vector HOG (DV-HOG) achieved equivalent or better detection accuracy compared with conventional HOG feature descriptors. A hardware implementation of DV-HOG occupies approximately 14.2 times smaller cell area than that of a conventional HOG implementation.
ER -