{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T17:40:32Z","timestamp":1791913232511,"version":"4.3.4"},"reference-count":39,"publisher":"MDPI AG","issue":"4","license":[{"start":{"date-parts":[[2019,2,22]],"date-time":"2019-02-22T00:00:00Z","timestamp":1550793600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["91738302"],"award-info":[{"award-number":["91738302"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["91438203"],"award-info":[{"award-number":["91438203"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Recently, extensive convolutional neural network (CNN)-based methods have been used in remote sensing applications, such as object detection and classification, and have achieved significant improvements in performance. Furthermore, there are a lot of hardware implementation demands for remote sensing real-time processing applications. However, the operation and storage processes in floating-point models hinder the deployment of networks in hardware implements with limited resource and power budgets, such as field-programmable gate arrays (FPGAs) and application-specific integrated circuits (ASICs). To solve this problem, this paper focuses on optimizing the hardware design of CNN with low bit-width integers by quantization. First, a symmetric quantization scheme-based hybrid-type inference method was proposed, which uses the low bit-width integer to replace floating-point precision. Then, a training approach for the quantized network is introduced to reduce accuracy degradation. Finally, a processing engine (PE) with a low bit-width is proposed to optimize the hardware design of FPGA for remote sensing image classification. Besides, a fused-layer PE is also presented for state-of-the-art CNNs equipped with Batch-Normalization and LeakyRelu. The experiments performed on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset using a graphics processing unit (GPU) demonstrate that the accuracy of 8-bit quantized model drops by about 1%, which is an acceptable accuracy loss. The accuracy result tested on FPGA is consistent with that of GPU. As for the resource consumptions of FPGA, the Look Up Table (LUT), Flip-flop (FF), Digital Signal Processor (DSP), and Block Random Access Memory (BRAM) are reduced by 46.21%, 43.84%, 45%, and 51%, respectively, compared with that of floating-point implementation.<\/jats:p>","DOI":"10.3390\/s19040924","type":"journal-article","created":{"date-parts":[[2019,2,22]],"date-time":"2019-02-22T03:49:44Z","timestamp":1550807384000},"page":"924","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":36,"title":["FPGA-Based Hybrid-Type Implementation of Quantized Neural Networks for Remote Sensing Applications"],"prefix":"10.3390","volume":"19","author":[{"given":"Xin","family":"Wei","sequence":"first","affiliation":[{"name":"Beijing Key Laboratory of Embedded Real-time Information Processing Technology, Beijing Institute of Technology, Beijing 100081, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7747-523X","authenticated-orcid":false,"given":"Wenchao","family":"Liu","sequence":"additional","affiliation":[{"name":"Beijing Key Laboratory of Embedded Real-time Information Processing Technology, Beijing Institute of Technology, Beijing 100081, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lei","family":"Chen","sequence":"additional","affiliation":[{"name":"Beijing Key Laboratory of Embedded Real-time Information Processing Technology, Beijing Institute of Technology, Beijing 100081, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5011-2161","authenticated-orcid":false,"given":"Long","family":"Ma","sequence":"additional","affiliation":[{"name":"School of Information Engineering, Zhengzhou University, Zhengzhou 450001, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"He","family":"Chen","sequence":"additional","affiliation":[{"name":"Beijing Key Laboratory of Embedded Real-time Information Processing Technology, Beijing Institute of Technology, Beijing 100081, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0443-1081","authenticated-orcid":false,"given":"Yin","family":"Zhuang","sequence":"additional","affiliation":[{"name":"School of Electronics Engineering and Computer Science, Peking University, Beijing 100087, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2019,2,22]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"258","DOI":"10.1109\/JOE.2017.2767106","article-title":"Ship Classification in TerraSAR-X Images With Convolutional Neural Networks","volume":"43","author":"Bentes","year":"2018","journal-title":"IEEE J. Ocean. Eng."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"937","DOI":"10.1109\/TGRS.2017.2756851","article-title":"Multisource Remote Sensing Data Classification Based on Convolutional Neural Network","volume":"56","author":"Xu","year":"2018","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Li, W., Dong, R., Fu, H., and Yu, a.L. (2018). Large-Scale Oil Palm Tree Detection from High-Resolution Satellite Images Using Two-Stage Convolutional Neural Networks. Remote Sens., 11.","DOI":"10.3390\/rs11010011"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Gong, Z., Zhong, P., Hu, W., and Hua, Y. (2019). Joint Learning of the Center Points and Deep Metrics for Land-Use Classification in Remote Sensing. Remote Sens., 11.","DOI":"10.3390\/rs11010076"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1293","DOI":"10.1109\/LGRS.2017.2708722","article-title":"M-FCN: Effective Fully Convolutional Network-Based Airplane Detection Framework","volume":"14","author":"Yang","year":"2017","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"937","DOI":"10.1109\/LGRS.2018.2813094","article-title":"Arbitrary-Oriented Ship Detection Framework in Optical Remote-Sensing Images","volume":"15","author":"Liu","year":"2018","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Liu, W., Ma, L., Wang, J., and Chen, H. (2018). Detection of Multiclass Objects in Optical Remote Sensing Images. IEEE Geosci. Remote Sens. Lett., 1\u20135.","DOI":"10.1109\/LGRS.2018.2882778"},{"key":"ref_8","unstructured":"Kamel, A., Maxime, P., Jocelyn, S., and Fran\u00e7ois, B. Personal communication."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1432","DOI":"10.1109\/TCSI.2016.2647322","article-title":"A Scalable Network-on-Chip Microprocessor With 2.5D Integrated Memory and Accelerator","volume":"64","author":"Sai","year":"2017","journal-title":"IEEE Trans. Circuits Syst. I Regul. Pap."},{"key":"ref_10","unstructured":"Chen, T., Du, Z., Sun, N., Jia, W., Wu, C., Chen, Y., and Temam, O. (2014, January 1\u20135). DianNao: A small-footprint high-throughput accelerator for ubiquitous machine-learning. Proceedings of the International Conference on Architectural Support for Programming Languages and Operating Systems, Salt Lake City, UT, USA."},{"key":"ref_11","unstructured":"Chen, Z., Di, W., Sun, J., Sun, G., Luo, G., and Cong, J. (2016, January 8\u201310). Energy-Efficient CNN Implementation on a Deeply Pipelined FPGA Cluster. Proceedings of the International Symposium on Low Power Electronics and Design, San Francisco, CA, USA."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Qiu, J., Song, S., Yu, W., Yang, H., Jie, W., Song, Y., Guo, K., Li, B., Zhou, E., and Yu, J. (2016, January 21\u201323). Going Deeper with Embedded FPGA Platform for Convolutional Neural Network. Proceedings of the Acm\/sigda International Symposium on Field-programmable Gate Arrays, Monterey, CA, USA.","DOI":"10.1145\/2847263.2847265"},{"key":"ref_13","unstructured":"Iandola, F.N., Han, S., Moskewicz, M.W., Ashraf, K., Dally, W.J., and Keutzer, K. (arXiv, 2016). SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5 MB model size, arXiv."},{"key":"ref_14","unstructured":"Krizhevsky, A., Sutskever, I., and Hinton, G.E. (2012, January 3\u20138). ImageNet Classification with Deep Convolutional Neural Networks. Proceedings of the International Conference on Neural Information Processing Systems, Harrahs and Harveys, Lake Tahoe, CA, USA."},{"key":"ref_15","unstructured":"Howard, A.G., Zhu, M., Bo, C., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., and Adam, H. (arXiv, 2017). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications, arXiv."},{"key":"ref_16","unstructured":"Simonyan, K., and Zisserman, A. (arXiv, 2014). Very Deep Convolutional Networks for Large-Scale Image Recognition, arXiv."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Zhang, X., Zhou, X., Lin, M., and Jian, S. (arXiv, 2017). ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices, arXiv.","DOI":"10.1109\/CVPR.2018.00716"},{"key":"ref_18","unstructured":"Liu, B., Min, W., Foroosh, H., Tappen, M., and Penksy, M. (2015, January 7\u201312). Sparse Convolutional Neural Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"303","DOI":"10.1007\/s11263-009-0275-4","article-title":"The Pascal Visual Object Classes (VOC) Challenge","volume":"88","author":"Everingham","year":"2009","journal-title":"Int. J. Comput. Vis."},{"key":"ref_20","unstructured":"Denil, M., Shakibi, B., Dinh, L., Ranzato, M.A., and Freitas, N.D. (2013, January 5\u201310). Predicting parameters in deep learning. Proceedings of the International Conference on Neural Information Processing Systems, Harrahs and Harveys, Lake Tahoe, CA, USA."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"1695","DOI":"10.1109\/LGRS.2018.2859024","article-title":"IORN: An Effective Remote Sensing Image Scene Classification Framework","volume":"15","author":"Wang","year":"2018","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_22","unstructured":"Gupta, S., Agrawal, A., Gopalakrishnan, K., and Narayanan, P. (2015, January 6\u201311). Deep Learning with Limited Numerical Precision. Proceedings of the International Conference on Machine Learning, Lille, France."},{"key":"ref_23","unstructured":"Gysel, P. (arXiv, 2016). Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks, arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Guadarrama, S., and Darrell, T. (arXiv, 2014). Caffe: Convolutional Architecture for Fast Feature Embedding, arXiv.","DOI":"10.1145\/2647868.2654889"},{"key":"ref_25","unstructured":"Courbariaux, M., Bengio, Y., and David, J.P. (arXiv, 2014). Training deep neural networks with low precision multiplications, arXiv."},{"key":"ref_26","unstructured":"Miyashita, D., Lee, E.H., and Murmann, B. (arXiv, 2016). Convolutional Neural Networks using Logarithmic Data Representation, arXiv."},{"key":"ref_27","unstructured":"Zhou, A., Yao, A., Guo, Y., Xu, L., and Chen, Y. (2017, January 6\u20139). Incremental Network Quantization: Towards Lossless CNNs with Low-precision Weights. Proceedings of the International Conference on Learning Representations, New Orleans, LA, USA."},{"key":"ref_28","unstructured":"Courbariaux, M., Bengio, Y., and David, J.P. (2015, January 7\u201312). BinaryConnect: Training Deep Neural Networks with binary weights during propagations. Proceedings of the International Conference on Neural Information Processing Systems, Montr\u00e9al, QC, Canada."},{"key":"ref_29","unstructured":"Li, F., and Liu, B. (arXiv, 2016). Ternary Weight Networks, arXiv."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Rastegari, M., Ordonez, V., Redmon, J., and Farhadi, A. (2016, January 8\u201316). XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks. Proceedings of the European Conference on Computer Vision, Amsterdam, The Netherlands.","DOI":"10.1007\/978-3-319-46493-0_32"},{"key":"ref_31","first-page":"3","article-title":"Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding","volume":"56","author":"Song","year":"2015","journal-title":"Fiber"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Jacob, B., Kligys, S., Chen, B., Zhu, M., Tang, M., Howard, A.G., Adam, H., and Kalenichenko, D. (2018, January 19\u201321). Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.","DOI":"10.1109\/CVPR.2018.00286"},{"key":"ref_33","unstructured":"Lei, C., Xin, W., Wenchao, L., He, C., and Liang, C. (2018, January 14\u201316). Hardware Implementation of Convolutional Neural Network Based Remote Sensing Image Classification Method. Proceedings of the 7th International Conference on Communications, Signal Processing, and Systems (CSPS), Dalian, China."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"228","DOI":"10.1117\/12.242059","article-title":"MSTAR extended operating conditions: a tutorial","volume":"2757","author":"Keydel","year":"1996","journal-title":"Proc. SPIE"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"539","DOI":"10.1109\/49.761034","article-title":"Analog-to-digital converter survey and analysis","volume":"17","author":"Walden","year":"1999","journal-title":"IEEE J. Sel. Areas Commun."},{"key":"ref_36","unstructured":"Ioffe, S., and Szegedy, C. (2015, January 6\u201311). Batch normalization: accelerating deep network training by reducing internal covariate shift. Proceedings of the International Conference on Machine Learning, Lille, France."},{"key":"ref_37","unstructured":"Andrew, L.M., Awni, Y.H., and Andrew, Y.N. (2013, January 16\u201321). Rectifier nonlinearities improve neural network acoustic models. Proceedings of the International Conference on Machine Learning, Atlanta, GA, USA."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"Lecun","year":"1998","journal-title":"Proc. IEEE"},{"key":"ref_39","unstructured":"Jake, B. Personal communication."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/19\/4\/924\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T12:34:02Z","timestamp":1760186042000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/19\/4\/924"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,2,22]]},"references-count":39,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2019,2]]}},"alternative-id":["s19040924"],"URL":"https:\/\/doi.org\/10.3390\/s19040924","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,2,22]]}}}