Efficient combination of Lidar intensity and 3D information by DNN for pedestrian recognition with high and low density 3D sensor

Urban trees are essential for sustaining biodiversity, mitigating climate impacts, and promoting public health. Among them, oak trees (Quercus spp.) hold particular ecological and cultural importance but are also associated with risks posed by the oak processionary caterpillar, an invasive pest with direct health implications. Effective monitoring and management of urban oaks require precise, scalable mapping tools that go beyond the limitations of traditional field surveys. This thesis develops a novel framework integrating airborne LiDAR, aerial photographs, and deep learning to detect and map individual oak trees across the Netherlands.<br/>The research first addresses the persistent challenge of spatial misalignment between multimodal datasets. A new tree-oriented matching approach was designed to align aerial photographs and LiDAR point clouds at the individual tree level, significantly improving registration compared to conventional image-based methods. Building on this foundation, a decision-level fusion strategy was introduced to combine detection outputs from aerial photographs and LiDAR data, leveraging their complementary strengths. This approach achieved higher detection accuracy than pixel-level fusion by mitigating residual misregistration errors.<br/>The thesis then advances species classification using side-view projections of LiDAR point clouds. By optimizing the number of projections and integrating intensity information, deep learning models demonstrated clear advantages over traditional classifiers, achieving robust performance even with low-density national LiDAR data. An EfficientNetV2-S architecture proved particularly effective for distinguishing English oak from morphologically similar species, with fine-tuning enabling strong model transferability across diverse urban environments.<br/>These methodological innovations were combined into a nationwide case study, identifying more than two million oak trees across 1,045 Dutch urban areas. Spatial analyses revealed distinct patterns of oak distribution in relation to land use and population density, highlighting areas of heightened exposure risk to oak processionary caterpillars.<br/>Overall, the thesis contributes a transferable and scalable deep learning framework for urban tree mapping that advances both ecological research and public health planning. By integrating multimodal airborne data with state-of-the-art machine learning, this work provides actionable tools for municipalities, urban planners, and biodiversity managers to design healthier, more resilient cities.<br/>

Lidar (Light detection and ranging) is a relatively new method for the acquisition of terrain surface. The Lidar system not only generates the 3-dimensional cloud of points with irregular spacing, but also gets the laser impulse reflection data. Algorithms and software used for the Lidar data are deal with the 3-dimensional cloud of points, and get the DTM (Digital Terrain Model). The algorithms apply in the laser impulse reflection data are little. According to the character of Lidar data, which is fused the 3-dimension information and intensity information in pixel-level, two new fusion filtering algorithms are proposed in this paper. These algorithms translate the intensity information of Lidar data into intensity image, and fused the 3- dimension information, mend the traditional image procession ways to deal with the Lidar data. All the algorithms mentioned are applied in Lidar data, and their results are compared in different evaluation parameters. The conclusion proved the improvement of these proposed algorithms in keeping the advantage of traditional filtering algorithms under the condition of preserving edges information of the intensity image.

Airborne lidar is a technology well-suited for mapping many forest attributes, including aboveground biomass (AGB) stocks and changes in selective logging in tropical forests. However, trade-offs still exist between lidar pulse density and accuracy of AGB estimates. We assessed the impacts of lidar pulse density on the estimation of AGB stocks and changes using airborne lidar and field plot data in a selectively logged tropical forest located near Paragominas, Pará, Brazil. Field-derived AGB was computed at 85 square 50 × 50 m plots in 2014. Lidar data were acquired in 2012 and 2014, and for each dataset the pulse density was subsampled from its original density of 13.8 and 37.5 pulses·m−2 to lower densities of 12, 10, 8, 6, 4, 2, 0.8, 0.6, 0.4 and 0.2 pulses·m−2. For each pulse density dataset, a power-law model was developed to estimate AGB stocks from lidar-derived mean height and corresponding changes between the years 2012 and 2014. We found that AGB change estimates at the plot level were only slightly affected by pulse density. However, at the landscape level we observed differences in estimated AGB change of &gt;20 Mg·ha−1 when pulse density decreased from 12 to 0.2 pulses·m−2. The effects of pulse density were more pronounced in areas of steep slope, especially when the digital terrain models (DTMs) used in the lidar derived forest height were created from reduced pulse density data. In particular, when the DTM from high pulse density in 2014 was used to derive the forest height from both years, the effects on forest height and the estimated AGB stock and changes did not exceed 20 Mg·ha−1. The results suggest that AGB change can be monitored in selective logging in tropical forests with reasonable accuracy and low cost with low pulse density lidar surveys if a baseline high-quality DTM is available from at least one lidar survey. We recommend the results of this study to be considered in developing projects and national level MRV systems for REDD+ emission reduction programs for tropical forests.

Lidar technology has become an important data source in 3D terrain modelling. In Spain, the National Plan for Aerial Orthophotography will soon release public low-density lidar data (0.5–1 pulses/m2) for most of the country territory. Taking advantage of this fact, this article experimentally assesses the possibility of classifying a rural landscape into eight classes using multitemporal and multidensity lidar data and analyses the effect of point density on classification accuracy. Two statistical methods (transformed divergence and the Jeffries–Matusita distance) were used to assess the possibility of discriminating the eight classes and to determine which data layers were best suited for classification purposes. The results showed that ‘dirt road’ cannot be discriminated from ‘bare earth’ and that the possibility of discriminating ‘bare earth’, ‘pavement’, and ‘low vegetation’ decreases when using densities below 4 pulses/m2. Two non-parametric tests, the Kruskal–Wallis test and the Friedman test, were used to strengthen the results by assessing their statistical significance. According to the results of the Kruskal–Wallis test, lidar point density does not significantly affect the classification, whereas the results of the Friedman test show that bands could be considered as the only parameter affecting the possibility of discriminating some of the classes, such as ‘high vegetation’. Finally, the J48 algorithm was used to perform cross-validation in order to obtain the most familiar quantitative values in the international literature (e.g. overall accuracy). Mean overall accuracy was around 85% when the eight classes were considered and increased up to 95% when ‘dirt road’ was disregarded.

Effect of dietary 25-hydroxycholecalciferol supplementation and high stocking density on performance, egg quality, and tibia quality in laying hens

Mediterranean dehesa ecosystems are highly valuable agroforestry systems from ecological, social and economic perspectives. Their structural characterization has traditionally relied on resource-intensive field inventories. This study assesses the applicability of low-density airborne LiDAR data from the Spanish National Aerial Orthophotography Plan (PNOA) for the automated morphological characterization of Quercus ilex dehesas. This novel workflow integrates the DBSCAN clustering algorithm for unsupervised segmentation of tree vegetation units and Concaveman for crown perimeter delineation and slicing using concave hulls. The technique was applied over 116 hectares in Santibáñez el Bajo (Cáceres), identifying 1254 vegetation units with 99.8% precision, 97.3% recall and an F-score of 98.5%. A field validation on 35 trees revealed strong agreement with the LiDAR-derived metrics, including crown diameter (R2 = 0.985; bias = −0.96 m) and total height (R2 = 0.955; bias = −0.34 m). Crown base height was overestimated (+0.77 m), leading to a 20.9% underestimation of crown volume, which was corrected using a regression model (R2 = 0.952). This methodology allows us to produce scalable, fully automated forest inventories across extensive Iberian dehesas with similar structural characteristics using publicly available LiDAR data, even with a six-year acquisition gap.

Using a combination of remote sensing data, Pinus pinaster Ait. and Pinus pinea L. were distinguished at individual tree level in mixed Mediterranean stands with over 95% accuracy. This approach is easily applicable over large areas, enhancing the economic value of non-wood forest products, stone pine nuts, and resin, and aiding forest managers to accurately predict this production. The discrimination of tree species at individual level in mixed Mediterranean forest based on remote sensing is a field which has gained greater importance. In these stands, the capacity to predict the quality and quantity of non-wood forest products is particularly important due to the very different goods the two species produce. To assess the potential of using low-density airborne LiDAR data combined with high-resolution Pleiades images to discriminate two different pine species in mixed Mediterranean forest (Pinus pinea L. and Pinus pinaster Ait.) at individual tree level. A Random Forest model was trained using plots from the pure stand dataset, determining which LiDAR and satellite variables allow us to obtain better discrimination between groups. The model constructed was then validated by classifying individuals in an independent set of pure and mixed stands. The model combining LiDAR and Pleiades data provided greater accuracy (83.3% and 63% in pure and mixed validation stands, respectively) than the models which only use one type of covariables. The automatic crown delineation tool developed allows two very similar species in mixed Mediterranean conifer forest to be discriminated using continuous spatial information at the surface: Pleiades images and open source LiDAR data. This approach is easily applicable over large areas, enhancing the economic value of non-wood forest products and aiding forest managers to accurately predict production.

High stocking density is used in aquaculture to optimize farming. However, such strategies can stress territorial fish by increasing aggression, thus harming the fish welfare and productive performance. Here, we tested the effect of long-term tactile body stimulation (TS) on alleviating the impact of high stocking density in Nile tilapia (Oreochromis niloticus). Ten males were grouped for 21 days into four treatments (10 replicates each one) combining TS with high (1.2 kg·m³) or low (0.6 kg·m³) densities: (1) without TS and high density; (2) without TS and low density; (3) with TS and high density; and (4) with TS and low density. A rectangular PVC frame fitted with vertical plastic sticks sided with silicone bristles was placed in the center of aquarium. The fish receive TS when passing through the bristles. An apparatus without bristles was used as control. We found higher cortisol levels under TS conditions. All parameters of productive performance (growth, food efficiency, and condition factor) were higher in the treatment with TS and low density, lower in the treatment without TS and high density, and similar between treatments high density with TS and low density without TS. The results suggest that TS improves productive performance of Nile tilapia, counteracting the negative effects of high stocking density.

Three-dimensional characterization of pine forest type and red-cockaded woodpecker habitat by small-footprint, discrete-return lidar

Abstract. Growing demand for an efficient land use above and below the ground is motivating cadastre and land management systems to move from traditional 2D systems toward three dimensional ones. Airborne laser technology offers direct acquisition of dense and accurate 3D data. In order to get 3D road this paper proposes a hierarchical algorithm to extract terrain point from LIDAR data. We stratify the raw LiDAR data according to the height, judge terrain points and non-terrain points by the connectivity. In the case of road network, it indicates the morphological characteristics of network structure with a certain length continuous strip and small difference in intensity. All these information, including elevation information, the intensity information, the morphological characteristics and other local features, are used for extracting the road network from DTM. Local morphological filtering method is implementing for finding clear boundaries and rich details of the road profile. Following the presentation of the algorithm results for this approach are shown and evaluated

Abstract. Automatic registration of multi-sensor data is a basic step in data fusion for photogrammetric and remote sensing applications. The effectiveness of intensity-based methods such as Mutual Information (MI) for automated registration of multi-sensor image has been previously reported for medical and remote sensing applications. In this paper, a new multivariable MI approach that exploits complementary information of inherently registered LiDAR DSM and intensity data to improve the robustness of registering optical imagery and LiDAR point cloud, is presented. LiDAR DSM and intensity information has been utilised in measuring the similarity of LiDAR and optical imagery via the Combined MI. An effective histogramming technique is adopted to facilitate estimation of a 3D probability density function (pdf). In addition, a local similarity measure is introduced to decrease the complexity of optimisation at higher dimensions and computation cost. Therefore, the reliability of registration is improved due to the use of redundant observations of similarity. The performance of the proposed method for registration of satellite and aerial images with LiDAR data in urban and rural areas is experimentally evaluated and the results obtained are discussed.

This dissertation addresses the problem of automated vector extraction from Airborne Laser Scanner (ALS) data in urban areas. The recent popularity of Geographic Information Systems (GIS) has stimulated research on automated object extraction in order to simplify the data acquisition and update process. By automatically generating GIS inputs and updates from a single data source, the cost of data acquisition and processing will be kept to a minimum. Compared with other remote sensing techniques, extraction of objects from ALS data is in its infancy. ALS sensor technology has evolved rapidly and now allows the acquisition of very dense point clouds in a short period of time. ALS data is unique in that it explicitly contains 3D information and is acquired from an active sensor. As such, there are several benefits that can be immediately realised by using an ALS only data approach; data acquisition will not be limited to daylight hours as with other sensors and accurate height information is contained in the data. This means that registration of different data sources is not required and as only one data source is used, acquisition costs are minimised. Apart from these facts, ALS data has some other unique properties that have not been utilised to their full potential. An ALS sensor can record both height and intensity information from multiple returns in a single swath. To date, all the available information has rarely been used. For example, the intensity of a laser return has regularly been dismissed as it was considered under-sampled and noisy. This data still contains usable information and until this information is used, the optimum object recognition results will not be achieved; thus the use of as much of this information as possible is a major focus of this thesis. As ALS is an explicit 3D data source, the early stages of development were primarily focused on topographic mapping of terrain in forested areas in order to generate Digital Terrain Models (DTMs). As sensor technology has improved, so has the achievable resolution of point clouds from ALS data, and methods to extract objects from stand-alone ALS data have emerged. Attempts have been made to create city models and maps from ALS data that included buildings, roads, trees and powerlines. Each of these spatial object types has unique attributes, which means that automatic map creation is not an easy task. For example, buildings in general are easily detected in ALS data but the building outline is not easily delineated. Another difficulty with building detection is the separation of buildings and bridges as they have many of the same properties as observed by an ALS system. Bridges can usually be found in a road network but road extraction techniques typically produce poor detection rates and often require existing data and / or user interaction in semi-automatic techniques. These simple examples highlight the complexity of automatically generating vectorised maps from ALS data or in fact any data source. In this thesis, new methods are presented for the automatic creation of vectorised maps from ALS data. A two-step processing paradigm is adopted for this purpose, namely the classification of the ALS data and the vectorisation of the classification results. A classification strategy is introduced that creates a hierarchy for object detection with respect to the ALS data itself. This approach develops an ontology between the spatial object classes and the ALS data. The hierarchical framework highlights the fact that one object might not be discernable within the data without considering another. New classification algorithms are then presented within this framework. Each algorithm attempts to exploit the attributes of the data that are consistent within the spatial object class being considered as described by the classification framework. New algorithms for the classification of roads, trees and powerlines are all introduced whilst an extension to existing building classification methods is presented. Once classification is complete, a vectorisation process specific to the task at hand can be employed to yield vectorised results. These developed vectorisation processes are new and include an algorithm that has been generalised to allow the vectorisation of thick lines in images by detecting the centreline, direction and width. The primary goal of this thesis is to present a framework of new algorithms that will allow automatic spatial object detection and vectorisation whilst providing results of an acceptable quality. The algorithms presented rely solely on ALS data and require minimal operator knowledge. Each algorithm has been designed to exploit the way in which the object exhibits itself in the data. The new algorithms are integrated into a software package called JTD (Join The Dots) that will facilitate the effective automatic processing of ALS data. The results of the new algorithms have been evaluated over four 2 x 2 km areas that have been sampled with medium resolution ALS data. The results for each area are displayed and analysed to show the applicability of the whole process in an exemplary way.

Cortisol level changes in response to stocking density in the early stages of rainbow trout were measured. Eggs were exposed to low, normal, and high (2.55, 5.10 and 7.65 eggs cm(-2)) densities during the incubation period. Cortisol of maternal origin was found in pre-fertilized eggs (5.09 ± 0.12 ng g(-1)) of rainbow trout. In newly fertilized eggs, resting Cortisol levels (3.68 ± 0.14 ng g(-1)) decreased to 0.58 ± 0.08, 0.60 ± 0.12, and 0.57 ± 0.16 ng g(-1) at low, normal and high densities by day 10 (organo-genesis), respectively. Resting Cortisol levels remained constant until the eyed stage (day 18). Then, Cortisol showed an increase at hatching to 1.16 ± 0.11, 1.20 ± 0.12, and 1.21 ± 0.14 ng g(-1) at low, normal, and high densities, respectively. The pattern of change in Cortisol level was similar in all three densities. Interrenal cells were observed in 1-day old alevins at all three densities. Hematopoietic tissue, renal tubules and nucleated red blood cells were clarified through the head part of kidney. Higher numbers and larger interrenal cells were observed at high-density groups. Chronic density stress test conducted on embryonic stages of rainbow trout revealed no differences in Cortisol levels, but had an effect on the abundance and size of the interrenal cells. Densities were equaled after hatching (200 alevins per replicate) to investigate the different densities of eggs on stress indices in rainbow trout alevins. An acute stress (air exposure of eggs for five minutes) was applied in three treatments two weeks after hatching, and samples were taken at 0, 1, 3, 6, and 24 hps (hours post stress). Cortisol content increased under low density in 1 hps and reached from 5.21 ± 0.13 ng g(-1) to 6.01 ± 0.18 ng g(-1) (P < 0.05). Cortisol levels increased under normal density in 1 hps from 6.03 ± 0.28 ng g(-1) to 10.84 ± 0.18 ng g(-1) (P < 0.05). In high density also Cortisol increased from 6.83 ± 0.23 ng g(-1) to 8.86 ± 0.26 ng g(-1) (P < 0.05). At 3 hps;, Cortisol level was returned to basal level under low (P > 0.05) and normal (P > 0.05) densities, but significantly decreased (P < 0.05) under high density. Results revealed that the Cortisol biosynthesis was observed in rainbow trout between eyeing to the hatching stage. An increase in the density of eggs until 7.65 egg cm(-2) impaired Cortisol secretion and feedback system in alevins. However, more studies are needed to identify the exact time for Cortisol synthesis ability from eyeing to hatching in this species.

In this study, we investigate the effect of local adaptation to developmental density on male mating success in laboratory populations of Drosophila melanogaster. Mating success is known to be influenced by body condition which can in turn be influenced by local adaptation. We test the hypothesis that males adapted to a given environment have higher mating success when assayed in that environment. We used males selected for adaptation to high larval density and their controls which are reared at low larval density. We grew assay males in low and high densities whereas the focal females (raised at low larval density) used for the experiment belonged to the common ancestor of selected and control populations. We considered selected males grown at high density and control males grown at low density as 'adapted'. Similarly, we considered selected males grown at low density and control males grown at high density as 'nonadapted'. Selected male belonging to a given treatment (larval density) was made to compete with control male of the same treatment for mating with ancestral female. We quantified components of reproductive fitness: mating latency, copulation duration, mating success and number of progeny sired by the 'adapted' and 'nonadapted' males. The results show that local adaptation does not lead to higher mating success in populations adapted to their own larval rearing environment.

Three-dimensional (3D) imaging systems are being researched extensively for purposes of sensing and visualization in fields as diverse as defense, medical, art, and entertainment. When compared to traditional 2D imaging techniques, 3D imaging offers advantages in ranging, robustness to scene occlusion, and target recognition performance. Amongst the myriad 3D imaging techniques, 3D multiperspective imaging technologies have received recent attention due to the technologies’ relatively low cost, scalability, and passive sensing capabilities. Multiperspective 3D imagers collect 3D scene information by recording 2D intensity information from multiple perspectives, thus retaining both ray intensity and angle information. Three novel developments in 3D sensing, imaging, and visualization systems are presented: 3D imaging with axially distributed sensing, 3D optical profilometry, and occluded 3D object tracking.