Local Maxima Detection Research Articles

The Acoustic Laboratory for Marine Applications (ALMA) applied to fluctuating environment analysis

The Acoustic Laboratory for Marine Applications (ALMA) has been used to address problems in underwater acoustics, such as sound propagation in fluctuating environments. In this work, data from the ALMA-2016 at-sea campaign are used to analyze the ocean fluctuation's influence on sound propagation in a shallow-water waveguide. The experiment took place in November 2016 on the continental shelf of the eastern coast of the island of Corsica. A source and a receiver array were 9.3 km apart in a nearly constant water depth of 100 m. A thermistor chain was moored near the source to monitor sound speed fluctuations. The source emitted a variety of signals from which the chirp (1–13 kHz) is used to extract the waveguide eigenrays. To do so, a time-domain beamforming is performed on the match-filtered received signals with an automatic detection of local maxima in the time of arrival/direction of arrival (TOA/DOA) domain. A 2 min acquisition period of more than 13 h duration shows significant fluctuations in eigenray TOAs/DOAs. Qualitative comparisons with synthetic signals obtained from simulations permit reproduction of the observed eigenray fluctuations without including range dependence of the sound-speed profile. In addition, the joint analysis of the probability density function of the normalized acoustic intensity and of the thermistor chain data highlights the time dependence of the received signal characteristics.

Multi-scale fusion public gathering recognition based on residual network

With the rapid economic development and high concentration of urban population, people’s income level and quality of life continue to improve, resulting in more and more crowded scenes caused by people going out. Especially in urban commercial centers, transportation hubs, sports venues during important events, tourist attractions, etc., crowd gatherings occur frequently. However, accidents involving crowd gatherings in public places occur frequently, causing heavy casualties and property losses. Therefore, for crowd recognition, this paper proposes a new method to accurately estimate the number of dense crowds. In this method, a density map with accurate pedestrian locations is first generated using the focal inverse distance transform and used as ground truth labels for network training. Then, a multi-scale feature fusion algorithm based on residual network is designed, combining spatial and channel attention mechanisms to improve the accuracy and stability of crowd density estimation. In dense crowds, the phenomenon of overlapping and occlusion of people is very common and serious, making it difficult for existing pedestrian detection methods to distinguish each individual and accurately count the flow of people. To solve this problem, this paper proposes a density map-based method that uses a local maximum detection strategy and a K-nearest neighbor algorithm to convert the density map into the corresponding dense head bounding box. This method can effectively reduce the impact of occlusion and improve the accuracy of people counting. In order to further improve the estimation accuracy, a pattern recognition density peak clustering algorithm is introduced to study the clustered crowds. By treating the head bounding box as an element point, the distance between each element point is calculated, and the density of each point is calculated. Then perform clustering to find the cluster center with the highest density in each class. Finally, by comparing the density of each cluster center with the corresponding density threshold and adopting the corresponding decision-making method, the accuracy of people counting is further improved.

Toward the Optimization of Mining Operations Using an Automatic Unmineable Inclusions Detection System for Bucket Wheel Excavator Collision Prevention: A Synthetic Study

This work introduces a methodology for the automatic unmineable inclusions detection and Bucket Wheel Excavator (BWE) collision prevention, using electromagnetic (EM) inspection and a fuzzy inference system. EM data are collected continuously ahead from the bucket wheel of a BWE and subjected to processing. Two distinct methodologies for data processing were developed and integrated into the MATLAB programming environment. The first approach, named “Simple Mode”, utilizes statistical process control to generate real-time alerts in the event of a potential collision involving the excavator’s bucket and hard rock inclusions. The advanced processing flow (“Advanced Mode”) requires accurate instrument positioning and data from successive EM scans. It incorporates techniques of local resistivity maxima detection (Position Prominence Index) as well as Neural Network-based Pattern Recognition (NNPR). A decision support process based on a Fuzzy Inference System (FIS) has been developed to assist BWE operators in avoiding collision when digging hard rock inclusions. The proposed methodology was extensively tested using synthetic EM data. Limited real data, acquired with a CMD2 (GF Instruments) EM instrument equipped with GPS, were used to control its efficiency. Increased accuracy in the automatic detection of unmineable inclusions was observed using the Advanced Mode. On the other hand, the Simple Mode processing technique offers the advantage of being independent of instrument positioning as well as it provides real-time inspection of the excavated mine slope. This work introduces a methodology for hard rock inclusion detection and can contribute to the optimization of mine operations by improving resource efficiency, safety, cost savings, and environmental sustainability.

Aircraft Detection in SAR Images Based on Peak Feature Fusion and Adaptive Deformable Network

Due to the unique imaging mechanism of synthetic aperture radar (SAR), targets in SAR images often shows complex scattering characteristics, including unclear contours, incomplete scattering spots, attitude sensitivity, etc. Automatic aircraft detection is still a great challenge in SAR images. To cope with these problems, a novel approach called adaptive deformable network (ADN) combined with peak feature fusion (PFF) is proposed for aircraft detection. The PFF is designed for taking full advantage of the strong scattering features of aircraft, which consists of peak feature extraction and fusion. To fully exploit the strong scattering features of the aircraft in SAR images, peak features are extracted via the Harris detector and the eight-domain pixel detection of local maxima. Then, the saliency of aircraft under multiple imaging conditions is enhanced by multi-channel blending. All the PFF-preprocessed images are fed into the ADN for training and testing. The core components of ADN contain an adaptive spatial feature fusion (ASFF) module and a deformable convolution module (DCM). ASFF is utilized to reconcile the inconsistency across different feature scales, raising the characterization capabilities of the feature pyramid and improving the detection performance of multi-scale aircraft further. DCM is introduced to determine the 2-D offsets of feature maps adaptively, improving the geometric modeling abilities of aircraft in various shapes. The well-designed ADN is established by combining the two modules to alleviate the problems of the multi-scale targets and attitude sensitivity. Extensive experiments are conducted on the GaoFen-3 (GF3) dataset to demonstrate the effectiveness of the PFF-ADN with an average precision of 89.34%, as well as an F1-score of 91.11%. Compared with other mainstream algorithms, the proposed approach achieves state-of-the-art performance.

Shallow-water waveguide acoustic analysis in a fluctuating environment.

The Acoustic Laboratory for Marine Applications (ALMA) is a deployable and autonomous acoustic system, designed by DGA Naval Systems, to address problems in underwater acoustics, such as sound propagation in fluctuating environments. In this article, data from the ALMA-2016 at-sea campaign are used to analyze the ocean fluctuation's influence on sound propagation in a shallow-water waveguide. The experiment took place on the continental shelf of the island of Corsica in November 2016. A source and a receiver array were 9.3 km apart in a nearly constant water depth of 100 m. The source emitted a variety of signals from which the chirp (1-13 kHz) is used to extract the waveguide eigenrays. To do so, a time-domain beamforming is performed on the match-filtered received signals with an automatic detection of local maxima in the time of arrival/direction of arrival (TOA/DOA) domain. A 2 min acquisition period of more than 13 h duration shows significant fluctuations in eigenray TOAs/DOAs. Qualitative comparisons with synthetic signals obtained from simulations in two and three dimensions permit reproduction of the observed eigenray fluctuations without including range dependence of the sound-speed profile.

DETECTION OF METHANE PLUMES IN HYPERSPECTRAL IMAGES FROM SENTINEL-5P BY COUPLING ANOMALY DETECTION AND PATTERN RECOGNITION

Abstract. Reducing methane emissions is essential to tackle climate change. Here, we address the problem of detecting large methane leaks using hyperspectral data from the Sentinel-5P satellite. For that we exploit the fine spectral sampling of Sentinel-5P data to detect methane absorption features visible in the shortwave infrared wavelength range (SWIR). Our method involves three separate steps: i) background subtraction, ii) detection of local maxima in the negative logarithmic spectrum of each pixel and iii) anomaly detection in the background-free image. In the first step, we remove the impact of the albedo using albedo maps and the impact of the atmosphere by using a principal component analysis (PCA) over a time series of past observations. In the second step, we count for each pixel the number of local maxima that correspond to a subset of local maxima in the methane absorption spectrum. This counting method allows us to set up a statistical a contrario test that controls the false alarm rate of our detections. In the last step we use an anomaly detector to isolate potential methane plumes and we intersect those potential plumes with what have been detected in the second step. This process strongly reduces the number of false alarms. We validate our method by comparing the detected plumes against a dataset of plumes manually annotated on the Sentinel-5P L2 methane product.

A non-convex regularization based line artefact quantification method in lung ultrasound imagery for pulmonary disease evaluation

Lung (Pulmonary) diseases are among the most severe health problems which cause multiple deaths (more than 100 thousand people in the UK) every year. Moreover, following the COVID-19 pandemic, analysis and diagnosis of pulmonary disease became even more crucial. The common feature in all clinical conditions, both local to the lungs [e.g., pneumonia, chronic obstructive pulmonary disease (COPD)] and those manifesting themselves in the lungs (e.g., kidney disease, COVID-19) is the presence in lung ultrasound (LUS) images of a variety of artefacts. This work presents a novel method for line artefacts quantification in LUS images of pulmonary disease patients by using a non-convex regularization method. We employ a simple local maxima detection technique in the Radon transform domain, associated with known clinical definitions of line artefacts. Notwithstanding its non-convex characteristics, the proposed technique is guaranteed to converge through our proposed Cauchy proximal splitting (CPS) method and accurately identifies both horizontal (pleural, sub-pleural, A-) and vertical (B- and Z-) line artefacts in LUS images. The proposed method includes a two-stage validation mechanism, which is performed in both Radon and image domains to reduce the number of false and missed detections.

Detection of Line Artifacts in Lung Ultrasound Images of COVID-19 Patients Via Nonconvex Regularization.

In this article, we present a novel method for line artifacts quantification in lung ultrasound (LUS) images of COVID-19 patients. We formulate this as a nonconvex regularization problem involving a sparsity-enforcing, Cauchy-based penalty function, and the inverse Radon transform. We employ a simple local maxima detection technique in the Radon transform domain, associated with known clinical definitions of line artifacts. Despite being nonconvex, the proposed technique is guaranteed to convergence through our proposed Cauchy proximal splitting (CPS) method, and accurately identifies both horizontal and vertical line artifacts in LUS images. To reduce the number of false and missed detection, our method includes a two-stage validation mechanism, which is performed in both Radon and image domains. We evaluate the performance of the proposed method in comparison to the current state-of-the-art B-line identification method, and show a considerable performance gain with 87% correctly detected B-lines in LUS images of nine COVID-19 patients.

An Automated Method for Developing a Catalog of Small Earthquakes Using Data of a Dense Seismic Array and Nearby Stations

AbstractWe propose a new automated procedure for using continuous seismic waveforms recorded by a dense array and its nearby regional stations for P-wave arrival identification, location, and magnitude estimation of small earthquakes. The method is illustrated with a one-day waveform dataset recorded by a dense array with 99 sensors near Anza, California, and 24 surrounding regional stations within 50 km of the dense array. We search a wide range of epicentral locations and apparent horizontal slowness values (0–15 s/km) in the 15–25 Hz range and time shift the dense array waveforms accordingly. For each location–slowness combination, the average neighboring station waveform similarity (avgCC) of station pairs <150 m apart is calculated for each nonoverlapping 0.5 s time window. Applying the local maximum detection algorithm gives 966 detections. Each detection has a best-fitting location–slowness combination with the largest avgCC. Of 331 detections with slowness <0.4 s/km, 324 (about six times the catalog events and 98% accuracy) are found to be earthquake P-wave arrivals. By associating the dense array P-wave arrivals and the P- and S-wave arrivals from the surrounding stations using a 1D velocity model, 197 detections (∼4 times of the catalog events) have well-estimated locations and magnitudes. Combining the small spacing of the array and the large aperture of the regional stations, the method achieves automated earthquake detection and location with high sensitivity in time and high resolution in space. Because no preknowledge of seismic-waveform features or local velocity model is required for the dense array, this automated algorithm can be robustly implemented in other locations.

Vibration Measurement Based on the Local Maximum Detection Algorithm for Laser Self-Mixing Interferometry

A novel laser self-mixing interferometry vibration measurement algorithm based on the local maximum detection technique is proposed, that can reconstruct micro vibrations rapidly and easily. In this article, the principles of the local maximum detection algorithm are analyzed in detail, and the process of solving the window function is emphasized. The major advantage of the method is that it does not involve any complicated calculations and removes the need for optical/electromechanical components. The validity of the presented method is confirmed by means of simulated signals and demonstrated via several experiments for harmonic and aleatory motion.

Wavelet-based clod segmentation on digital elevation models of a soil surface with or without furrows

Soil surface roughness is a key factor for our understanding and modelling of geomorphologic processes related to exchanges of soil, water and gas. It has an impact on soil properties and tillage outcome. Soil surface roughness can be characterized both globally and locally. The interest of clod segmentation is to allow for both characterizations. Segmenting clods on a digital elevation model (DEM) of a soil surface is a complex problem because soil surfaces are complex surfaces of several level of roughness and because considering elevations results in smooth and poorly contrasted images. However, a DEM in 3 dimensions gathers more information than a profile of 1 dimension or an image of 2 dimensions.Multiresolution analysis has shown interest for roughness analysis of complex surfaces. We have used it to introduce a new approach for soil roughness analysis and to lay the foundations for clod segmentation. In this paper, we propose a complete wavelet-based approach for accurate clod contour delineation. It relies on several steps: detecting clods on the surface approximations by a supervised detection of local maxima, validating and merging the detections by shape and overlap tests, delineating the clod contours by intersecting locally the soil surface elevations with the estimated plane of the clod base and validating the contours by detecting and correcting the wrong patterns, with statistical pattern recognition. This segmentation method was evaluated in several roughness conditions, made in the laboratory, by comparison with other segmentation method. An indicator of goodness of agreement was introduced for this purpose.This wavelet-based segmentation method showed robustness to the presence of furrows and to the smoothing by rainfall and showed ability to retrieve clod diameters.

Ejection Wave Segmentation for Contact-Free Heart Rate Estimation from Ballistocardiographic Signals.

We present a new algorithm for peak detection in ballistocardiographic (BCG) signals and use it for heart rate estimation. Systolic complexes of the BCG signal are enhanced and coarse heart beat locations estimated. Ejection waves I, J and K are detected simultaneously around coarse locations, only using detection of local maxima and weighted summation of peak heights. Due to a lack of reference BCG annotations, the algorithm's performance is assessed by using the detected peaks for heart rate estimation. On a dataset acquired with a pneumatic BCG system, we evaluate the heart rate estimation performance and compare the introduced algorithm against other methods found in literature. The dataset is gathered from 42 patients in a clinical environment and provides low-quality signals taken from a realistic scenario. With a mean absolute percentage error of 2.58 % at 65 % coverage, the presented method is on par with the best-performing state-of-the-art algorithms investigated. Limits of agreement (5th/95th percentiles) in a comparison with ECG-based heart rate measurements lie within P5 = -3.63 and P95 = 5.78 beat/min.

INDIVIDUAL TREE DETECTION FROM UAV LIDAR DATA IN A MIXED SPECIES WOODLAND

Abstract. The study evaluates five existing segmentation algorithms to determine the method most suitable for individual tree detection across a species-diverse forest: raster-based region growing, local maxima centroidal Voronoi tessellation, point-cloud level region growing, marker controlled watershed and continuously adaptive mean shift. Each of the methods has been tested twice over one mixed and five single species plots: with their parameters set as constant and with the parameters calibrated for every plot. Overall, continuous adaptive mean shift performs best across all the plots with average F-score of 0.9 with fine-tuned parameters and 0.802 with parameters held at constant. Raster-based algorithms tend to achieve higher scores in coniferous plots, due to the clearly discernible tops, which significantly aid the detection of local maxima. Their performance is also highly dependent on the moving size window used to detect the local maxima, which ideally should be readjusted for every plot. Crown overlap, suppressed and leaning trees are the most likely sources of error for all the algorithms tested.

The Algorithm of Local Modulus Maximum Detection Based on B-spline Wavelet

At the time of fault, the system signal performances singularity, and the fault state can be diagnosed effectively through effective analysis and identification of singular signals. In this paper, for the problem that the singular signal modulus maximum point detection is inaccurate and the polarity is confused due to the improper selection of wavelet basis, the algorithm of local modulus maximum detection based on b-spline wavelet is proposed. B-spline scaling function is constructed by interpolation formula, the filter coefficients of b-spline wavelet function is calculated, in the process of signal decomposition of b-spline wavelet function and the effect of endpoint effect on signal singularity is removed, the signal local modulus maximum in the wavelet decomposition coefficient is extracted to accurately determine the time when the fault occurs. The result of simulation and experimental shows that the algorithm can not only detect the location of the maximum value of the singular point modulus accurately, but also distinguish the polarity of the modulus maximum and have good anti noise characteristics.

Ідентифікація та оцінювання висот дерев сосни звичайної у насадженнях Чорнобильської зони відчуження стереофотограмметричним способом

Розглянуто можливості використання аерофотозйомки за допомогою безпілотного літального апарату та подальшого оброблення отриманих зображень стереофотограмметричними підходами для ідентифікації дерев сосни звичайної та оцінки їх висоти у зімкнутих деревостанах способом порівняння даних, отриманих для кругової пробної площі внаслідок аерофотозйомки та в польових умовах лазерно-оптичними далекомірами, що використані як контрольні значення. Встановлено можливість ідентифікації верхівок стовбурів дерев сосни звичайної за допомогою алгоритмів пошуку максимумів растрів у програмних продуктах ГІС. Підтверджено неможливість детектування дерев нижніх елементів ярусу шляхом застосування радіусів пошуку максимумів малих розмірів (0,5 м), що спричиняє помилкову ідентифікацію значної кількості локальних максимумів (бокових гілок) як окремих дерев. Збільшення розмірів радіусу фільтрації виявлення максимальних значень растра цифрової моделі висоти крони досліджуваного деревостану до 1 м дало змогу з достатньою точністю оцінити висоту (стандартне відхилення 0,44 м) 27 дерев у верхній частині лісового пологу. Це свідчить про перспективність використання перелічених підходів для встановлення лісотаксаційних параметрів насаджень не лише з науковою метою, а й під час проведення лісовпорядкування і лісоінвентаризації, оскільки верхня висота деревостану має тісні функціональні зв'язки з іншими таксаційними ознаками.

Mapping the height and spatial cover of features beneath the forest canopy at small-scales using airborne scanning discrete return Lidar

The objective of the current study was to develop methods for estimating the height and horizontal coverage of the forest understorey using airborne Lidar data in three managed pine plantation forest typical of the south eastern USA. The current project demonstrates a two-step approach applied automatically across a given study site extent. The first operation divided the study site extent into a regularly spaced grid (25 × 25 m) and identified the potential height range of the main Loblolly pine canopy layer for each grid-cell through aggregating Lidar return height measurements into a ‘stack’ of vertical height bins describing the frequency of returns by height. Once height bins were created, the resulting vertical distributions were smoothed with a regression curve line function and the main canopy vertical layer was identified through the detection of local maxima and minima. The second operation sub-divided the 25 × 25 m grid-cell into 1 × 1 m horizontal grid, for which height-bin stacks were created for each cell. Vertical features below the main canopy were then identified at this scale in the same manner as in the previous step, and classified as understorey features if they were lower in height than the 25 × 25 m estimate of the main canopy layer. The heights of the tallest understorey and sub-canopy layers were kept, and used to produce a rasterized map of the understorey layer height at the 1 × 1 m scale. Lidar derived estimates of the 25 × 25 m lowest vertical extent of the coniferous canopy correlated highly with field data (R2 0.87; RMSE 2.1 m). Estimates of understorey horizontal cover ranged from R2 0.80 to 0.90 (RMSE 6.6–11.7%), and maximum understorey layer height ranged from R2 0.69 to 0.80 (RMSE 1.6–3.4 m) for the three study sites. The automated method deployed within the current study proved sufficient in determining the presence and absence of vegetation and artificial structures within the understorey portion of the current forest context, in addition to height and horizontal cover to a reasonable accuracy. Issues were encountered within older stands (e.g. more than 30 years old) where understorey deciduous vegetation layers intersected with the coniferous canopy layer, resulting in an underestimation of sub-dominant heights.

PolSAR Ship Detection Using Local Scattering Mechanism Difference Based on Regression Kernel

In this letter, the local scattering mechanism difference based on regression kernel (LSMDRK) is developed as a discriminative feature for ship detection. The LSMDRK measures the scattering mechanism dissimilarity of a center pixel to its neighboring pixels. A ship detection scheme is proposed based on the LSMDRK. The detection scheme consists of two stages. In the feature extraction stage, polarimetric target decomposition is required to improve the discriminative ability of the descriptor. In the detection stage, a saliency detection strategy is utilized to construct the saliency map. Then, local maximum detection is employed. Finally, an adaptive threshold method is designed to achieve the final detection. The effectiveness of the detection scheme is validated by a RADARSAT-2 data set. Experimental results demonstrate that the proposed method can acquire a better detection on weak targets and has much less false alarms than some classical detection methods.

Comparison of UAS photogrammetric products for tree detection and characterization of coniferous stands

ABSTRACTThe use of Unmanned Aerial Systems (UAS) opens a new era for remote sensing and forest management, which requires accurate and regular quantification of resources. In this study, we propose a comprehensive workflow to detect trees and assess forest attributes in the particular context of coniferous stands in transformation from even-aged to uneven-aged stands, using UAS imagery, from data acquisition to model construction. We implement a local maxima detection to identify the tree tops, based on a fixed-radius moving window in a Canopy Height Model (CHM) and images produced from UAS surveys. To compare the contribution of different photogrammetric products, we analysed the local maxima detected from the CHM, from three image types (individual rectified and ortho-rectified images and ortho-mosaic) and from a combination of both CHM and images. The local maxima detection gave promising results, with low omission and true-positive rates of up to 89.2%. A filtering process of false positives was implemented, using a supervised classification which decreased the false positives up to 2.6%. Based on the local maxima combined with an area-based approach, we constructed models to assess top height (R 2: 83%, root mean square error [RMSE]: 5.7%), number of stems (R 2: 71%, RMSE: 28.3%), basal area (R 2: 70%, RMSE: 16.2%), volume (R 2: 69%, RMSE: 20.1%), and individual tree height (R 2: 70%, RMSE: 7.2%). Despite a suboptimal data acquisition, our simple and flexible method has yielded good results and shows great potential for application.

Cyclic sources extraction from complex multiple-component vibration signal via periodically time varying filter

AbstractThe problem of local damage detection is widely discussed in the literature. There are many methods which can be applied, however there is still a need for new techniques addressing specific diagnostic issues. In particular, the case of complex multiple-component vibration signal is a challenging problem. In this paper we focus on such a problem related to a gearbox operating in industrial conditions. Our method consists of several stages. First we transform signal to time-frequency domain using spectrogram. Then for each frequency bin we apply a novel procedure which indicates location of cyclic impulses in given time series. This algorithm is based on the periodically distributed local maxima detection and quantification of their significance. Such procedure requires a priori known fault frequency. If the machine might reveal multiple fault, the procedure has to be calculated separately for each fault frequency. A time-varying filter is designed using the indicated local maxima comprised in the score matrix. Then, signals representing each fault frequency are obtained using inverse short-time Fourier transform algorithm. The method is illustrated with application to simulated and real data from complex mining machine - heavy duty gearbox.

PALM TREE DETECTION USING CIRCULAR AUTOCORRELATION OF POLAR SHAPE MATRIX

Abstract. Palm trees play an important role as they are widely used in a variety of products including oil and bio-fuel. Increasing demand and growing cultivation have created a necessity in planned farming and the monitoring different aspects like inventory keeping, health, size etc. The large cultivation regions of palm trees motivate the use of remote sensing to produce such data. This study proposes an object detection methodology on the aerial images, using shape feature for detecting and counting palm trees, which can support an inventory. The study uses circular autocorrelation of the polar shape matrix representation of an image, as the shape feature, and the linear support vector machine to standardize and reduce dimensions of the feature. Finally, the study uses local maximum detection algorithm on the spatial distribution of standardized feature to detect palm trees. The method was applied to 8 images chosen from different tough scenarios and it performed on average with an accuracy of 84% and 76.1%, despite being subjected to different challenging conditions in the chosen test images.