{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T13:00:04Z","timestamp":1790600404482,"version":"4.1.0"},"reference-count":45,"publisher":"Acoustical Society of America (ASA)","issue":"3","funder":[{"DOI":"10.13039\/501100010903","name":"Key Programme","doi-asserted-by":"publisher","award":["62071390"],"award-info":[{"award-number":["62071390"]}],"id":[{"id":"10.13039\/501100010903","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["pubs.aip.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,9,1]]},"abstract":"<jats:p>Reliable detection and recovery of multiple frequency line structures in Lofargrams are crucial for passive underwater acoustic target analysis. Although deep neural networks have shown promising performance for this task, most existing methods rely heavily on dense pixel-level annotations. To address this limitation, we propose a weakly supervised unified framework for multiple frequency line detection and recovery. Specifically, line detection is conducted under coarse region-level supervision, whereas line recovery is achieved through gradient-based visualization in a weakly supervised manner without pixel-level masks. To ensure that the recovered lines faithfully preserve fine-grained dynamic characteristics, such as continuity and fluctuations, we further introduce a recovery loss that enforces curvature smoothness and structural consistency. Extensive experiments show that the proposed multiple DeepLofargram achieves state-of-the-art performance on multiple coexisting frequency lines with fragmented and fluctuating structures under low-signal-to-noise conditions.<\/jats:p>","DOI":"10.1121\/10.0046669","type":"journal-article","created":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T12:25:49Z","timestamp":1790598349000},"page":"2608-2626","update-policy":"https:\/\/doi.org\/10.1063\/aip-crossmark-policy-page","source":"Crossref","is-referenced-by-count":0,"title":["Multiple DeepLofargram: Weakly supervised detection and recovery of multiple dim frequency lines for underwater acoustic targets"],"prefix":"10.1121","volume":"160","author":[{"given":"Ziyuan","family":"Xiao","sequence":"first","affiliation":[{"name":"School of Marine Science and Technology, Northwestern Polytechnical University 1 , Xi'an 710072,","place":["China"]},{"name":"Shaanxi Key Laboratory of Underwater Information Technology 2 , Xi'an 710072,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yutao","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Marine Science and Technology, Northwestern Polytechnical University 1 , Xi'an 710072,","place":["China"]},{"name":"Shaanxi Key Laboratory of Underwater Information Technology 2 , Xi'an 710072,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaojin","family":"Sun","sequence":"additional","affiliation":[{"name":"School of Marine Science and Technology, Northwestern Polytechnical University 1 , Xi'an 710072,","place":["China"]},{"name":"Shaanxi Key Laboratory of Underwater Information Technology 2 , Xi'an 710072,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9713-1821","authenticated-orcid":false,"given":"Yina","family":"Han","sequence":"additional","affiliation":[{"name":"School of Marine Science and Technology, Northwestern Polytechnical University 1 , Xi'an 710072,","place":["China"]},{"name":"Shaanxi Key Laboratory of Underwater Information Technology 2 , Xi'an 710072,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"231","published-online":{"date-parts":[[2026,9,28]]},"reference":[{"key":"2026092808254203200_c1","first-page":"561","article-title":"An image processing approach to frequency tracking (application to sonar data)","year":"1992"},{"issue":"1","key":"2026092808254203200_c2","doi-asserted-by":"publisher","first-page":"118","DOI":"10.1121\/1.428344","article-title":"Radiated noise characteristics of a modern cargo ship","volume":"107","year":"2000","journal-title":"J. Acoust. Soc. Am."},{"key":"2026092808254203200_c3","first-page":"1","article-title":"Improving vision transformers by revisiting high-frequency components","volume-title":"Computer Vision \u2013 ECCV 2022","year":"2022"},{"key":"2026092808254203200_c4","first-page":"111","article-title":"A theoretical analysis of feature pooling in visual recognition","year":"2010"},{"issue":"3","key":"2026092808254203200_c5","doi-asserted-by":"publisher","first-page":"1638","DOI":"10.1121\/1.2951592","article-title":"A sawtooth waveform inspired pitch estimator for speech and music","volume":"124","year":"2008","journal-title":"J. Acoust. Soc. Am."},{"issue":"10","key":"2026092808254203200_c6","doi-asserted-by":"publisher","first-page":"2586","DOI":"10.1109\/78.640725","article-title":"Characterization of signals by the ridges of their wavelet transforms","volume":"45","year":"1997","journal-title":"IEEE Trans. Signal Process."},{"issue":"4","key":"2026092808254203200_c7","doi-asserted-by":"publisher","first-page":"1917","DOI":"10.1121\/1.1458024","article-title":"Yin, a fundamental frequency estimator for speech and music","volume":"111","year":"2002","journal-title":"J. Acoust. Soc. Am."},{"issue":"1","key":"2026092808254203200_c8","doi-asserted-by":"publisher","first-page":"37","DOI":"10.1016\/0167-8655(95)00088-7","article-title":"An approach to detect lofar lines","volume":"17","year":"1996","journal-title":"Pattern Recognit. Lett."},{"issue":"1","key":"2026092808254203200_c9","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1186\/s13634-024-01181-9","article-title":"QiandaoEar22: A high-quality noise dataset for identifying specific ship from multiple underwater acoustic targets using ship-radiated noise","volume":"2024","year":"2024","journal-title":"EURASIP J. Adv. Signal Process."},{"key":"2026092808254203200_c10","first-page":"1440","article-title":"Fast r-cnn","year":"2015"},{"issue":"4","key":"2026092808254203200_c11","doi-asserted-by":"publisher","first-page":"2182","DOI":"10.1121\/10.0002172","article-title":"Deeplofargram: A deep learning based fluctuating dim frequency line detection and recovery","volume":"148","year":"2020","journal-title":"J. Acoust. Soc. Am."},{"key":"2026092808254203200_c12","doi-asserted-by":"publisher","first-page":"117252","DOI":"10.1016\/j.oceaneng.2024.117252","article-title":"A survey on machine learning in ship radiated noise","volume":"298","year":"2024","journal-title":"Ocean Eng."},{"issue":"5","key":"2026092808254203200_c13","doi-asserted-by":"publisher","first-page":"3204","DOI":"10.1121\/10.0034360","article-title":"Frequency line detection in spectrograms using a deep neural network with attention","volume":"156","year":"2024","journal-title":"J. Acoust. Soc. Am."},{"issue":"1","key":"2026092808254203200_c14","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1121\/1.1916342","article-title":"The sound spectrograph","volume":"18","year":"1946","journal-title":"J. Acoust. Soc. Am."},{"issue":"5","key":"2026092808254203200_c15","doi-asserted-by":"publisher","first-page":"1396","DOI":"10.1016\/j.patcog.2012.11.009","article-title":"On the detection of tracks in spectrogram images","volume":"46","year":"2013","journal-title":"Pattern Recognit."},{"key":"2026092808254203200_c16","first-page":"2643","article-title":"BBAM: Bounding box attribution map for weakly supervised semantic and instance segmentation","year":"2021"},{"issue":"13","key":"2026092808254203200_c17","doi-asserted-by":"publisher","first-page":"3268","DOI":"10.3390\/rs15133268","article-title":"Joint detection and reconstruction of weak spectral lines under non-Gaussian impulsive noise with deep learning","volume":"15","year":"2023","journal-title":"Remote Sens."},{"key":"2026092808254203200_c18","doi-asserted-by":"publisher","first-page":"3968","DOI":"10.1121\/10.0026438","article-title":"Enhanced noise-resilience in passive tone detection via broad-receptive field complex-valued convolutional neural networks","volume":"155","year":"2024","journal-title":"J. Acoust. Soc. Am."},{"key":"2026092808254203200_c19","first-page":"2980","article-title":"Focal loss for dense object detection","year":"2017"},{"key":"2026092808254203200_c20","first-page":"740","article-title":"Microsoft coco: Common objects in context","year":"2014"},{"issue":"22","key":"2026092808254203200_c21","doi-asserted-by":"publisher","first-page":"4866","DOI":"10.3390\/s19224866","article-title":"A sensing and tracking algorithm for multiple frequency line components in underwater acoustic signals","volume":"19","year":"2019","journal-title":"Sensors"},{"key":"2026092808254203200_c22","first-page":"659","article-title":"PYIN: A fundamental frequency estimator using probabilistic threshold distributions","year":"2014"},{"key":"2026092808254203200_c23","first-page":"4","article-title":"A new tracker for multiple frequency line","year":"2001"},{"issue":"2","key":"2026092808254203200_c24","doi-asserted-by":"publisher","first-page":"439","DOI":"10.1109\/TAES.2003.1207256","article-title":"Frequency line tracking using hmm-based schemes [passive sonar]","volume":"39","year":"2003","journal-title":"IEEE Trans. Aerosp. Electron. Syst."},{"issue":"13","key":"2026092808254203200_c25","doi-asserted-by":"publisher","first-page":"85","DOI":"10.19651\/j.cnki.emt.2106635","article-title":"An improved spectrum line extraction method based on u-net network","volume":"44","year":"2021","journal-title":"Electron. Meas. Technol."},{"key":"2026092808254203200_c26","unstructured":"Pei, M., Deng, L., and Chen, Y. (2022). \u201cResearch on underwater target line spectrum detection technology based on deep learning,\u201d Master thesis, Changzhou University, Changzhou, Jiangsu, China."},{"issue":"5","key":"2026092808254203200_c27","doi-asserted-by":"publisher","first-page":"591","DOI":"10.1109\/TIT.1974.1055282","article-title":"Single tone parameter estimation from discrete-time observations","volume":"20","year":"1974","journal-title":"IEEE Trans. Inf. Theory"},{"key":"2026092808254203200_c28","first-page":"1","article-title":"New passive acoustic monitoring in Monterey Bay National Marine Sanctuary","year":"2016"},{"issue":"5","key":"2026092808254203200_c29","doi-asserted-by":"publisher","first-page":"1018","DOI":"10.1109\/TAC.1981.1102775","article-title":"Aspects of dynamic programming in signal and image processing","volume":"26","year":"1981","journal-title":"IEEE Trans. Autom. Control"},{"key":"2026092808254203200_c30","first-page":"618","article-title":"Grad-CAM: Visual explanations from deep networks via gradient-based localization","year":"2017"},{"key":"2026092808254203200_c31","doi-asserted-by":"publisher","first-page":"699","DOI":"10.15949\/j.cnki.0371-0025.2023.04.008","article-title":"Line spectrum extraction of underwater acoustic target using deep convolution network and adaptive enhancement learning","volume":"48","year":"2023","journal-title":"Acta Acust."},{"key":"2026092808254203200_c32","article-title":"Smoothgrad: Removing noise by adding noise","year":"2017"},{"key":"2026092808254203200_c33","unstructured":"Springenberg, J. T., Dosovitskiy, A., Brox, T., and Riedmiller, M. (2014). \u201cStriving for simplicity: The all convolutional net,\u201d arXiv:1412.6806."},{"issue":"1","key":"2026092808254203200_c34","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1109\/MCOM.2009.4752682","article-title":"Underwater acoustic communication channels: Propagation models and statistical characterization","volume":"47","year":"2009","journal-title":"IEEE Commun. Mag."},{"issue":"24","key":"2026092808254203200_c35","doi-asserted-by":"publisher","first-page":"6434","DOI":"10.1109\/TSP.2018.2877176","article-title":"Phase-continuous frequency line track-before-detect of a tone with slow frequency variation","volume":"66","year":"2018","journal-title":"IEEE Trans. Signal Process."},{"key":"2026092808254203200_c36","volume-title":"Sonar Technology","year":"2010"},{"issue":"4","key":"2026092808254203200_c37","doi-asserted-by":"publisher","first-page":"540","DOI":"10.1109\/48.895362","article-title":"Optimal tonal detectors based on the power spectrum","volume":"25","year":"2000","journal-title":"IEEE J. Ocean. Eng."},{"issue":"1","key":"2026092808254203200_c38","doi-asserted-by":"publisher","first-page":"8901565","DOI":"10.1155\/2021\/8901565","article-title":"Ship radiated noise recognition technology based on ML-DS decision fusion","volume":"2021","year":"2021","journal-title":"Comput. Intell. Neurosci."},{"key":"2026092808254203200_c39","first-page":"579","article-title":"Visual transformers: Where do transformers really belong in vision models?","year":"2021"},{"issue":"6","key":"2026092808254203200_c40","doi-asserted-by":"publisher","first-page":"4870","DOI":"10.1121\/10.0041865","article-title":"Accurate Deeplofargram: Accurate dynamic dim frequency line recovery for underwater acoustic target via waveform diffusion model","volume":"158","year":"2025","journal-title":"J. Acoust. Soc. Am."},{"issue":"1","key":"2026092808254203200_c41","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1121\/1.3664053","article-title":"Properties of underwater acoustic communication channels in shallow water","volume":"131","year":"2012","journal-title":"J. Acoust. Soc. Am."},{"issue":"3","key":"2026092808254203200_c42","doi-asserted-by":"publisher","first-page":"240","DOI":"10.1109\/TASSP.1978.1163074","article-title":"Adaptive enhancement of multiple sinusoids in uncorrelated noise","volume":"26","year":"1978","journal-title":"IEEE Trans. Acoust. Speech. Signal Process."},{"key":"2026092808254203200_c43","first-page":"818","article-title":"Visualizing and understanding convolutional networks","volume-title":"European Conference on Computer Vision (ECCV), Zurich, Switzerland","year":"2014"},{"key":"2026092808254203200_c44","first-page":"3791","article-title":"Weakly supervised instance segmentation using class peak response","year":"2018"},{"issue":"3","key":"2026092808254203200_c45","doi-asserted-by":"publisher","first-page":"328","DOI":"10.3390\/jmse10030328","article-title":"Underwater sound characteristics of a ship with controllable pitch propeller","volume":"10","year":"2022","journal-title":"JMSE"}],"container-title":["The Journal of the Acoustical Society of America"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.aip.org\/asa\/jasa\/article-pdf\/160\/3\/2608\/21154035\/2608_1_10.0046669.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.aip.org\/asa\/jasa\/article-pdf\/160\/3\/2608\/21154035\/2608_1_10.0046669.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T12:25:52Z","timestamp":1790598352000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.aip.org\/jasa\/article\/160\/3\/2608\/3405449\/Multiple-DeepLofargram-Weakly-supervised-detection"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,1]]},"references-count":45,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,9,1]]}},"URL":"https:\/\/doi.org\/10.1121\/10.0046669","relation":{},"ISSN":["1520-8524"],"issn-type":[{"value":"1520-8524","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,9]]},"published":{"date-parts":[[2026,9,1]]}}}