{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,3,26]],"date-time":"2024-03-26T00:35:25Z","timestamp":1711413325421},"reference-count":27,"publisher":"MIT Press","issue":"4","content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2024,3,21]]},"abstract":"Abstract<\/jats:title>\n Recent advancements in deep learning have achieved significant progress by increasing the number of parameters in a given model. However, this comes at the cost of computing resources, prompting researchers to explore model compression techniques that reduce the number of parameters while maintaining or even improving performance. Convolutional neural networks (CNN) have been recognized as more efficient and effective than fully connected (FC) networks. We propose a column row convolutional neural network (CRCNN) in this letter that applies 1D convolution to image data, significantly reducing the number of learning parameters and operational steps. The CRCNN uses column and row local receptive fields to perform data abstraction, concatenating each direction's feature before connecting it to an FC layer. Experimental results demonstrate that the CRCNN maintains comparable accuracy while reducing the number of parameters and compared to prior work. Moreover, the CRCNN is employed for one-class anomaly detection, demonstrating its feasibility for various applications.<\/jats:p>","DOI":"10.1162\/neco_a_01653","type":"journal-article","created":{"date-parts":[[2024,3,8]],"date-time":"2024-03-08T20:57:20Z","timestamp":1709931440000},"page":"744-758","update-policy":"http:\/\/dx.doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":0,"title":["Column Row Convolutional Neural Network: Reducing Parameters for Efficient Image Processing"],"prefix":"10.1162","volume":"36","author":[{"given":"Seongil","family":"Im","sequence":"first","affiliation":[{"name":"Center for Opto-Electronic Materials and Devices, Korea Institute of Science and Technology, Seoul, 02792 Republic of Korea"},{"name":"Department of Chemical and Biomolecular Engineering, Yonsei University, Seoul, 03722 Republic of Korea seongil.im@kist.re.kr"}]},{"given":"Jae-Seung","family":"Jeong","sequence":"additional","affiliation":[{"name":"Center for Opto-Electronic Materials and Devices, Korea Institute of Science and Technology, Seoul, 02792 Republic of Korea wotmd104@kist.re.kr"}]},{"given":"Junseo","family":"Lee","sequence":"additional","affiliation":[{"name":"Center for Opto-Electronic Materials and Devices, Korea Institute of Science and Technology, Seoul, 02792 Republic of Korea"},{"name":"Department of Electrical and Computer Engineering, Korea University, Seoul, 02841 Republic of Korea leejunseo97@kist.re.kr"}]},{"given":"Changhwan","family":"Shin","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, Korea University, Seoul, 02841 Republic of Korea cshin@korea.ac.kr"}]},{"given":"Jeong Ho","family":"Cho","sequence":"additional","affiliation":[{"name":"Department of Chemical and Biomolecular Engineering, Yonsei University, Seoul, 03722 Republic of Korea jhcho94@yonsei.ac.kr"}]},{"given":"Hyunsu","family":"Ju","sequence":"additional","affiliation":[{"name":"Center for Opto-Electronic Materials and Devices, Korea Institute of Science and Technology (KIST), Seoul, 02792 Republic of Korea hyunsuju@kist.re.kr"}]}],"member":"281","published-online":{"date-parts":[[2024,3,21]]},"reference":[{"issue":"3","key":"2024032521540728200_B1","doi-asserted-by":"publisher","first-page":"252","DOI":"10.1109\/TSMCC.2002.804448","article-title":"Gaussian-based edge-detection methods: A survey","volume":"32","author":"Basu","year":"2002","journal-title":"IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews)"},{"key":"2024032521540728200_B2","article-title":"Anomaly detection using one-class neural networks","author":"Chalapathy","year":"2018"},{"key":"2024032521540728200_B3","article-title":"A survey of model compression and acceleration for deep neural networks","author":"Cheng","year":"2017"},{"issue":"1","key":"2024032521540728200_B4","doi-asserted-by":"publisher","first-page":"659","DOI":"10.1109\/JSEN.2021.3129323","article-title":"A novel learning strategy for the trade-off between accuracy and computational cost: A touch modalities classification case study","volume":"22","author":"Gianoglio","year":"2021","journal-title":"IEEE Sensors Journal"},{"key":"2024032521540728200_B5","doi-asserted-by":"publisher","first-page":"1789","DOI":"10.1007\/s11263-021-01453-z","article-title":"Knowledge distillation: A survey","volume":"129","author":"Gou","year":"2021","journal-title":"International Journal of Computer Vision"},{"issue":"5786","key":"2024032521540728200_B6","doi-asserted-by":"publisher","first-page":"504","DOI":"10.1126\/science.1127647","article-title":"Reducing the dimensionality of data with neural networks","volume":"313","author":"Hinton","year":"2006","journal-title":"Science"},{"key":"2024032521540728200_B7","article-title":"Mobilenets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017"},{"issue":"6","key":"2024032521540728200_B8","doi-asserted-by":"publisher","first-page":"1734","DOI":"10.3390\/s20061734","article-title":"Comparison of CNN algorithms on hyperspectral image classification in agricultural lands","volume":"20","author":"Hsieh","year":"2020","journal-title":"Sensors"},{"issue":"1","key":"2024032521540728200_B9","doi-asserted-by":"publisher","first-page":"106","DOI":"10.1113\/jphysiol.1962.sp006837","article-title":"Receptive fields, binocular interaction and functional architecture in the cat's visual cortex","volume":"160","author":"Hubel","year":"1962","journal-title":"Journal of Physiology"},{"key":"2024032521540728200_B10","doi-asserted-by":"publisher","first-page":"107398","DOI":"10.1016\/j.ymssp.2020.107398","article-title":"1D convolutional neural networks and applications: A survey","volume":"151","author":"Kiranyaz","year":"2021","journal-title":"Mechanical Systems and Signal Processing"},{"key":"2024032521540728200_B11","article-title":"The MNIST database of handwritten digits","author":"LeCun","year":"1998"},{"issue":"11","key":"2024032521540728200_B12","doi-asserted-by":"publisher","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":"Proceedings of the IEEE"},{"issue":"12","key":"2024032521540728200_B13","doi-asserted-by":"publisher","first-page":"124704","DOI":"10.1063\/5.0025881","article-title":"One-dimensional convolutional neural network (1D-CNN) image reconstruction for electrical impedance tomography","volume":"91","author":"Li","year":"2020","journal-title":"Review of Scientific Instruments"},{"key":"2024032521540728200_B14","article-title":"Why are convolutional nets more sample-efficient than fully-connected nets?","author":"Li","year":"2020"},{"issue":"10","key":"2024032521540728200_B15","doi-asserted-by":"publisher","first-page":"2531","DOI":"10.1109\/TED.2010.2062187","article-title":"Cross-point memory array without cell selectors: Device characteristics and data storage pattern dependencies","volume":"57","author":"Liang","year":"2010","journal-title":"IEEE Transactions on Electron Devices"},{"key":"2024032521540728200_B16","doi-asserted-by":"crossref","DOI":"10.1109\/ICDM.2008.17","article-title":"Isolation forest","volume-title":"Proceedings of the Eighth IEEE International Conference on Data Mining","author":"Liu","year":"2008"},{"key":"2024032521540728200_B17","article-title":"A robust ensemble-deep learning model for COVID-19 diagnosis based on an integrated CT scan images database","volume-title":"Proceedings of the IIE Annual Conference","author":"Maftouni","year":"2021"},{"key":"2024032521540728200_B18","article-title":"Adversarial autoencoders","author":"Makhzani","year":"2015"},{"issue":"3","key":"2024032521540728200_B19","doi-asserted-by":"publisher","first-page":"1065","DOI":"10.1214\/aoms\/1177704472","article-title":"On estimation of a probability density function and mode","volume":"33","author":"Parzen","year":"1962","journal-title":"Annals of Mathematical Statistics"},{"key":"2024032521540728200_B20","doi-asserted-by":"crossref","first-page":"591","DOI":"10.1016\/B978-0-12-802232-0.00022-0","article-title":"Quality evaluation and control of potato chips","volume-title":"Computer vision technology for food quality evaluation","author":"Pedreschi","year":"2016"},{"key":"2024032521540728200_B21","article-title":"Model compression via distillation and quantization","author":"Polino","year":"2018"},{"key":"2024032521540728200_B22","article-title":"Deep one-class classification","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Ruff","year":"2018"},{"key":"2024032521540728200_B23","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-59050-9_12","article-title":"Unsupervised anomaly detection with generative adversarial networks to guide marker discovery","volume-title":"Proceedings of the 25th International Conference on Information Processing in Medical Imaging","author":"Schlegl","year":"2017"},{"key":"2024032521540728200_B24","article-title":"Design implications of memristor-based RRAM cross-point structures","volume-title":"Proceedings of the 2011 Design, Automation and Test in Europe Conference","author":"Xu","year":"2011"},{"key":"2024032521540728200_B25","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2017.15","article-title":"On compressing deep models by low rank and sparse decomposition","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Yu","year":"2017"},{"issue":"3","key":"2024032521540728200_B26","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1145\/3446776","article-title":"Understanding deep learning (still) requires rethinking generalization","volume":"64","author":"Zhang","year":"2021","journal-title":"Communications of the ACM"},{"key":"2024032521540728200_B27","article-title":"To prune, or not to prune: Exploring the efficacy of pruning for model compression","author":"Zhu","year":"2017"}],"container-title":["Neural Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/neco\/article-pdf\/36\/4\/744\/2351746\/neco_a_01653.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/neco\/article-pdf\/36\/4\/744\/2351746\/neco_a_01653.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,25]],"date-time":"2024-03-25T21:54:37Z","timestamp":1711403677000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/neco\/article\/36\/4\/744\/119785\/Column-Row-Convolutional-Neural-Network-Reducing"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,3,21]]},"references-count":27,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2024,3,21]]},"published-print":{"date-parts":[[2024,3,21]]}},"URL":"https:\/\/doi.org\/10.1162\/neco_a_01653","relation":{},"ISSN":["0899-7667","1530-888X"],"issn-type":[{"value":"0899-7667","type":"print"},{"value":"1530-888X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2024,4]]},"published":{"date-parts":[[2024,3,21]]}}}