Low-contrast Objects Research Articles

Self-healing spiral phase contrast imaging

Spiral phase contrast imaging alleviates the information load by extracting the geometric features of objects and is one of the most representative branches of instant imaging processing. The self-healing capacity of edge detectors can enhance their robustness to obstacles in practical applications. Here, a self-healing spiral phase contrast imaging scheme is proposed and experimentally demonstrated by a liquid crystal edge detector combining a spiral phase, an axicon phase, and a lens phase. The spiral phase is encoded into a liquid crystal by photopatterning. Self-healing contrast imaging is characterized by a series of edge images of both high-contrast amplitude-type and low-contrast phase-type objects. This work extends the self-healing capacity of these detectors to instant imaging processing and paves the way for optical applications with self-healing features.

3DSAM-adapter: Holistic adaptation of SAM from 2D to 3D for promptable tumor segmentation

Despite that the segment anything model (SAM) achieved impressive results on general-purpose semantic segmentation with strong generalization ability on daily images, its demonstrated performance on medical image segmentation is less precise and unstable, especially when dealing with tumor segmentation tasks that involve objects of small sizes, irregular shapes, and low contrast. Notably, the original SAM architecture is designed for 2D natural images and, therefore would not be able to extract the 3D spatial information from volumetric medical data effectively. In this paper, we propose a novel adaptation method for transferring SAM from 2D to 3D for promptable medical image segmentation. Through a holistically designed scheme for architecture modification, we transfer the SAM to support volumetric inputs while retaining the majority of its pre-trained parameters for reuse. The fine-tuning process is conducted in a parameter-efficient manner, wherein most of the pre-trained parameters remain frozen, and only a few lightweight spatial adapters are introduced and tuned. Regardless of the domain gap between natural and medical data and the disparity in the spatial arrangement between 2D and 3D, the transformer trained on natural images can effectively capture the spatial patterns present in volumetric medical images with only lightweight adaptations. We conduct experiments on four open-source tumor segmentation datasets, and with a single click prompt, our model can outperform domain state-of-the-art medical image segmentation models and interactive segmentation models. We also compared our adaptation method with existing popular adapters and observed significant performance improvement on most datasets. Our code and models are available at: https://github.com/med-air/3DSAM-adapter

Detection of low-contrast objects by electrical prospecting methods

Many geological problems cannot be solved by electrical prospecting methods. Often the reason for this is a slight difference in the electrical resistivity of target objects from that of the host medium. At the same time, the use of the transverse magnetic (TM) type field makes it possible to identify low-contrast objects and delineate their boundaries with high accuracy. TM-type prospecting is more efficient since its signal is much more strongly affected by changes in resistivity and other electrodynamic parameters. The article examines one of the difficult cases – a search for kimberlite pipes in Yakutia. Such objects differ very little from the host medium in terms of the horizontal resistivity. Exploration works to search for kimberlite pipes in Yakutia, carried out by conventional electrical prospecting methods, have many problems. The article considers the results of successful TM-type prospecting field surveys where kimberlite pipes stood out very prominently. Also presented are the results of modeling an even more complex situation in which the pipes are located at depths of over 140 m and overlain by traps; there are many objects in the area that create additional anomalies.

Near‐Infrared Metalens Empowered Dual‐Mode High Resolution and Large FOV Microscope

AbstractThe spiral phase contrast microscope can clearly distinguish the morphological information of the low contrast objects (i.e., biological samples) owing to the isotropic edge‐enhancement effect, while the bright field microscope can image the overall morphology of amplitude objects. However, the imaging resolution, magnification, and field of view of conventional spiral phase contrast microscopes based on 4f filtering configuration are limited by the system's complexity. Here, compact dual‐mode microscopes working at near‐infrared using the engineered metalens are reported, which can be tuned between the spiral phase contrast imaging and bright field imaging by polarization control. The metalens combine the high‐resolution objective lens and polarization‐controlled phase filter into a single‐layer nanofins array. Two infinity‐corrected microscope systems are demonstrated to achieve subwavelength resolution (0.7λ), large magnification (58X), and large field of view (600 × 800 µm). Unstained onion epidermal is imaged by the microscope to show the dual‐mode imaging ability for the biological sample. Finally, a singlet dual‐mode microscope system is demonstrated to show the edge‐detection application for industrial standards. These results can open new opportunities in applications of biological imaging, industrial machine vision, and semiconductor inspection.

Method for controlling the radiation pattern of an underground radiation antenna

Currently, communication and signaling means of underground structures are implemented by wire lines. In addition, radio communications of various frequency ranges are used, operating within line of sight and having the disadvantage of interrupting radio communications in emergency situations. Low-speed broadcast systems using ultra-low frequencies are also used. This work shows the possibility of using subsurface georadar technology to organize a faster circular warning system for underground structures. Existing GPR instruments using ultra-wideband pulses in the 1 to 50 MHz band reach depths of several hundred meters. In this case, control of the radiation pattern of the underground radiation antenna is possible both in the horizontal and azimuthal planes. Modern methods of controlling antenna-feeder devices used in georadar are shown, allowing to improve user qualities for various applications. With the development of these methods, additional areas of application of these devices appear: in underground radio communication systems, for identifying low-contrast objects, for ground penetrating radar from aircraft. Based on the method for controlling the directional pattern of underground radiation antennas in the horizontal plane for stationary or specialized devices, a new method has been developed to control the main lobe of the directional pattern in the azimuthal plane of the transmitting antenna. A method has been proposed for controlling the radiation pattern in various technical implementations, which allows changing the position of the radiation pattern both discretely and smoothly. Examples of using the developed method for controlling the radiation pattern of an underground radiation antenna for the implementation of geophysical instruments of various types have been shown. The possibilities of locating underground radiation antennas in various soils have been proposed when using them in circular warning systems for underground structures.

Channelized hotelling observer-based low-contrast detectability on the ACR CT accreditation phantom: Part II. Repeatability study.

Objective and quantitative evaluation for low-contrast detectability that correlates with human observer performance is lacking for routine CT quality control testing. Channelized Hotelling observer (CHO) is considered a strong candidate to fill the need but has long been deemed impractical to implement due to its requirement of a large number of repeated scans in order to provide accurate and precise estimates of index of detectability (d'). In our previous work, we optimized a CHO model observer on the American College of Radiology (ACR) CT accreditation phantom and achieved accurate measurement of d' with only 1-3 repeat scans. In this work, we aim to validate the repeatability of the proposed CHO-based low-contrast evaluation on four scanner models using the ACR CT accreditation phantom. The repeatability test was performed on four different scanners from two major CT manufacturers: Siemens Force and Alpha; Canon Prism and Prime SP. An ACR CT phantom was scanned 10 times, each time after repositioning of the phantom. For each repositioning, 3 repeated scans were acquired at 24, 12, and 6 mGy on all four scanner models. CHO was applied at the measured dose levels for different low-contrast object sizes (4-6mm). The CHO was also applied to images created using deep learning-based reconstructions on Canon Prism and to four different scan/reconstruction modes on the Siemens Alpha, a photon-counting-detector (PCD)-CT. The repeatability was evaluated by the probability that a measurement would fall within the±15% tolerance (P<15% ). With the CHO setting optimized for the ACR phantom and the use of 3 repeated scans and 9 non-overlapping slices per scan, the CHO measurement could provide high repeatability with P<15% of 98.8%-99.9% at 12 mGy with IR reconstruction on all four scanners. On scanner A, P<15% were 91.5%-99.9% at the three dose levels and for all three object sizes while the numbers were 93.6%-99.998% on scanner B. P<15% were 96.5%-97.2% for the two deep learning reconstructions and 97.0%-99.97% for the four scan/reconstruction modes on the PCD-CT. The CHO provided highly repeatable measurements with over 95% probability that a CHO measurement would lie within the±15% tolerance for most of the dose levels and object sizes on the ACR phantom. The repeatability was maintained when the CHO was applied to images created with a commercial deep learning-based reconstruction and various scan/reconstruction modes on a PCD-CT. This study demonstrates that practical implementation of CHO for routine quality control and performance evaluation is feasible.

Deep Generative Adversarial Reinforcement Learning for Semi-Supervised Segmentation of Low-Contrast and Small Objects in Medical Images.

Deep reinforcement learning (DRL) has demonstrated impressive performance in medical image segmentation, particularly for low-contrast and small medical objects. However, current DRL-based segmentation methods face limitations due to the optimization of error propagation in two separate stages and the need for a significant amount of labeled data. In this paper, we propose a novel deep generative adversarial reinforcement learning (DGARL) approach that, for the first time, enables end-to-end semi-supervised medical image segmentation in the DRL domain. DGARL ingeniously establishes a pipeline that integrates DRL and generative adversarial networks (GANs) to optimize both detection and segmentation tasks holistically while mutually enhancing each other. Specifically, DGARL introduces two innovative components to facilitate this integration in semi-supervised settings. First, a task-joint GAN with two discriminators links the detection results to the GAN's segmentation performance evaluation, allowing simultaneous joint evaluation and feedback. This ensures that DRL and GAN can be directly optimized based on each other's results. Second, a bidirectional exploration DRL integrates backward exploration and forward exploration to ensure the DRL agent explores the correct direction when forward exploration is disabled due to lack of explicit rewards. This mitigates the issue of unlabeled data being unable to provide rewards and rendering DRL unexplorable. Comprehensive experiments on three generalization datasets, comprising a total of 640 patients, demonstrate that our novel DGARL achieves 85.02% Dice and improves at least 1.91% for brain tumors, achieves 73.18% Dice and improves at least 4.28% for liver tumors, and achieves 70.85% Dice and improves at least 2.73% for pancreas compared to the ten most recent advanced methods, our results attest to the superiority of DGARL. Code is available at GitHub.

A full-scale hierarchical encoder-decoder network with cascading edge-prior for infrared and visible image fusion

Most existing deep learning-based infrared and visible image fusion methods always fail to consider the full-scale long-range correlation and the prior knowledge, resulting in the fused images with low-contrast salient objects and blurred edge details. To overcome these drawbacks, a full-scale hierarchical encoder-decoder network with cascading edge-prior for infrared and visible image fusion is proposed. First, a top-down encoder extracts the hierarchical representations from source image. Then, to inject edge priors into the network and capture the progressive semantic correlations, a triple fusion mechanism is proposed including edge image fusion based on maximum fusion strategy, single-scale shallow layer fusion and full-scale semantic layer fusion based on dual-attention fusion (DAF) strategy. The fused full-scale semantic features (F2SF) are obtained by capturing the long-range affinities of the full-scale. At the same time, a cascading edge-prior branch (CEPB) is designed to embed the fused edge knowledge into fused single-scale shallow features, jointly guiding the decoder to focus on abundant details layer-by-layer on the basis of F2SF, thus recovering the edge and texture details of the fused image well. Finally, a novel loss function consisting of SSIM, intensity and edge loss is constructed to further maintain the network with better edge representation and reconstruction capability. Compared with existing state-of-the-art fusion methods, the proposed method has better performance in terms of both visual evaluation and objective evaluation on public datasets. The source code is available at https://github.com/lxq-jnu/FSFusion.

Segmentation of Images Used in Unmanned Aerial Vehicles Navigation Systems

The paper presents the results of the study of a two-stage procedure for selecting a reference object in the current image formed by a correlation-extreme system used for autonomous navigation of unmanned aerial vehicles. The aim of this paper is to theoretically evaluate the probability of selecting low-dimensional low-contrast objects in the segmented current image according to the proposed two-stage procedure. To achieve this goal, the problem of segmentation of images of the sighting surface and subsequent selection of the reference object in the presence of heterogeneous objects differing in brightness and area characteristics is solved. The most significant result is the justification of application of two-stage procedure of selection of the reference object in the current image by brightness and area parameters using the set thresholds. The significance of the obtained results consists in establishing the dependence of the probability of correct selection of the reference object on the noise level of the current images. It is shown that the probability of correct selection of the object in the image is a function of the threshold value and can be maximised by choosing its value. This approach allows to consider the influence of various factors leading to image noise on the quality of images formed by the navigation system. It is shown that when noise distorts more than 31% of the image pixels, the proposed two-stage procedure allows to ensure the selection of the reference object in the image with a probability not lower than 0.9.

Optimized Dropkey-Based Grad-CAM: Toward Accurate Image Feature Localization.

Regarding the interpretable techniques in the field of image recognition, Grad-CAM is widely used for feature localization in images to reflect the logical decision-making information behind the neural network due to its high applicability. However, extensive experimentation on a customized dataset revealed that the deep convolutional neural network (CNN) model based on Gradient-weighted Class Activation Mapping (Grad-CAM) technology cannot effectively resist the interference of large-scale noise. In this article, an optimization of the deep CNN model was proposed by incorporating the Dropkey and Dropout (as a comparison) algorithm. Compared with Grad-CAM, the improved Grad-CAM based on Dropkey applies an attention mechanism to the feature map before calculating the gradient, which can introduce randomness and eliminate some areas by applying a mask to the attention score. Experimental results show that the optimized Grad-CAM deep CNN model based on the Dropkey algorithm can effectively resist large-scale noise interference and achieve accurate localization of image features. For instance, under the interference of a noise variance of 0.6, the Dropkey-enhanced ResNet50 model achieves a confidence level of 0.878 in predicting results, while the other two models exhibit confidence levels of 0.766 and 0.481, respectively. Moreover, it exhibits excellent performance in visualizing tasks related to image features such as distortion, low contrast, and small object characteristics. Furthermore, it has promising prospects in practical computer vision applications. For instance, in the field of autonomous driving, it can assist in verifying whether deep learning models accurately understand and process crucial objects, road signs, pedestrians, or other elements in the environment.

Three-dimensional CT imaging in extensor tendons using deep learning reconstruction: optimal reconstruction parameters and the influence of dose.

The purpose of this study was to assess the optimal reconstruction parameters and the influence of tube current in extensor tendons three-dimensional computed tomography (3D CT) using deep learning reconstruction, using iterative reconstruction as a reference. In the phantom study, a cylindrical phantom with a 3mm rod simulated an extensor tendon was used. The phantom images were acquired at tube current of 50, 100, 150, 200, and 250 mA. In the clinical study, CT scans of hand tendons were performed on nine hands from eight patients. All images were reconstructed using advanced intelligent clear-IQ engine (AiCE) parameters (body, body sharp, brain CTA, and brain LCD) and adaptive iterative dose reduction three dimensional (AIDR 3D). The objective image quality for tendon detectability was evaluated by calculating the low-contrast object specific contrast-to-noise ratio (CNRLO) in the phantom study and CNR and coefficient of variation (CV) in the clinical study. In the phantom study, CNRLO (at 200 mA) of AiCE parameters (body, body sharp, brain CTA, and brain LCD) and AIDR 3D were 5.2, 5.3, 5.3, 5.8, and 5.0, respectively. In the clinical study, AiCE brain CTA was higher CNR and lower CV values compared to other reconstruction parameters. AiCE without dose reduction may be an effective strategy for further improving the image quality of extensor tendons 3D CT. Our study suggests that the AiCE brain CTA is more suitable for extensor tendons 3D CT compared to other AiCE parameters.

Surface topography detection based on an optical differential metasurface.

Surface topography detection can extract critical characteristics from objects, playing an important role in target identification and precision measurement. Here, an optical method with the advantages of low power consumption, high speed, and simple devices is proposed to realize the surface topography detection of low-contrast phase objects. By constructing reflected light paths, a metasurface can perform spatial differential operation via receiving the light directly reflected from a target. Therefore, our scheme is experimentally demonstrated as having remarkable universality, which can be used not only for opaque objects, but also for transparent pure phase objects. It provides a new, to the best of our knowledge, application for optical differential metasurfaces in precise detection of microscale surface topography.

The stabilization of visibility for sequentially presented, low-contrast objects: Experiments and neural field model.

In any environment, events transpire in temporal sequences. The general principle governing such sequences is that each instance of the event is influenced by its predecessors. It is shown here that this principle is true for a fundamental aspect of visual perception: visibility. A series of nine psychophysical experiments and associated neural dynamic simulations provide evidence that two non-stimulus factors, self-excitation and short-term memory, stabilize the visibility of a simple low-contrast object (a line segment) as it moves over a sequence of unpredictable locations. Stabilization was indicated by the very low probability of visible-to-invisible switches, and dependence on preceding visibility states was indicated by hysteresis as the contrast of the object was gradually decreased or increased. The contribution of self-excitation to stabilization was indicated by increased visible-to-invisible switching (decreased hysteresis) following adaptation of the visibility state, and the contribution of memory to stabilization was indicated by visibility "bridging" long blank intervals separating each relocation of the object. Because of the unpredictability of the relocations of the object, its visibility at one location pre-shapes visibility at its next location via persisting subthreshold activation of detectors surrounding the low-contrast object. All effects were modeled, including contributions from adaptation and recurrent inhibition, with a single set of parameter values.

Evaluation of the feasibility of a multisource CBCT for maxillofacial imaging

Objective. The aim of this study was to investigate the feasibility of improving the image quality and accuracy of cone beam computed tomography (CBCT) by replacing the conventional wide cone angle x-ray tube with a distributed x-ray source array positioned in the axial direction. Approach. The multisource CBCT (ms-CBCT) design was experimentally simulated using a benchtop scanner with a carbon nanotube x-ray tube and a flat-panel detector. The source was collimated and translated in the axial direction to simulate a source array with a reduced cone angle for each beam. An adjacent scatter ratio subtraction (ASRS) method was implemented for residual scatter reduction. Several phantoms were imaged using the ms-CBCT and conventional CBCT configurations under otherwise similar conditions. The Requirements of the ms-CBCT design on the x-ray source and detector were evaluated. Main results. Compared to the conventional CBCT, the ms-CBCT design with 8 sources and ASRS significantly improved the image quality and accuracy, including: (1) reducing the cupping artifact from 15% to 3.5%; (2) reducing the spatial nonuniformity of the CT Hounsfield unit values from 38.0 to 9.2; (3) improving the contrast-to-noise ratio of the low contrast objects (acrylic and low density polyethylene inserts) against the water-equivalent background by ∼20% and (4) reducing the root-mean-square error of the HU values by 70%, from 420.1 to 124.4. The imaging dose and scanning time used by the current clinical CBCT for maxillofacial imaging can be achieved by current source and detector technologies. Significance. The ms-CBCT design significantly reduces the scatter and improves the image quality and accuracy compared to the conventional CBCT.

A Method of Superimposing Subapertures for Shark–Hartmann Wave Front Sensing with Faint Objects

Abstract For the astronomical adaptive optics systems, the observation condition is always in a regime of low photon flux. We present a method of superimposing subapertures for the Shark–Hartmann wave front sensing with faint objects. This approach significantly improves the signal-to-noise ratio (S/N) of the spot pattern by superimposing adjacent subapertures, thus circumventing the invalid wave front sensing with faint object, enhancing the accuracy of the wave front detection. In addition, we demonstrate that the validity of this approach is not an incidental result under a specific condition, which can be explained theoretically by exploiting the atmospheric turbulence model. Then we validate the approach with simulations and experimental realization under weak-light and bright-light conditions. Results show that the proposed subaperture superposition approach can achieve accurate wave front reconstruction for low S/N or faint object even in cases beyond the ability of conventional methods. It brings out a new idea, besides long exposure and variable sampling, that comes with a tradeoff between temporal sampling frequency and spatial resolution. In addition, this method is also valid for extended objects; we preliminarily validate this method on a 600 mm aperture solar telescope, achieving stable closed-loop control with lower-contrast objects.

The Performance of Sensitivity-Maps Method in Reconstructing Low Contrast and Multi-Contrast Objects for Microwave Imaging Applications

The microwave imaging system for breast tumor/cancer detection requires high sensitivity to detect abnormal tissue that has little contrast in high-density breasts. This paper proposes a qualitative microwave imaging system simulation based on inverse scattering using the sensitivity-maps method. This method utilizes two measurement types for system calibration: a reference object as a scatterer-free background and a calibration object to obtain the system's impulse response. The object under test (OUT) consists of an object with low dielectric contrast and a phantom with multiple low dielectric contrasts (multi-contrast). Reconstruction is carried out on three types of S-parameter measurement data, namely S_11,〖 S〗_21, and a combination of both. S-parameters are measured at several frequencies, which are 3, 10, 14, 15, 16, 20 GHz, and the combination of all those frequencies (multifrequency). Reconstructed images show that the system is capable of reconstructing dielectric objects accurately. Quantitatively, the results show that the multifrequency S_21 measurement yields the best image quality with relative root mean squared error (RRMSE) values of 0.1272 and structural similarity index (SSIM) of 0.9076. The designed imaging system also successfully reconstructs multi-contrast phantom accurately with RRMSE of 0.1434 and SSIM of 0.4609.

Noise reduction performance of a deep learning-based reconstruction in brain computed tomography images acquired with organ-based tube current modulation.

We aimed to evaluate the image quality of brain computed tomography (CT) images reconstructed using deep learning-based reconstruction (DLR) in organ-based tube current modulation (OB-TCM) acquisition. An anthropomorphic head phantom and a cylindrical low-contrast phantom were scanned at the standard dose level for adult brain CT in axial volume acquisition without OB-TCM. Moreover, image acquisition with OB-TCM was performed. The radiation dose on the eye lens was measured using a scintillation fibre-optic dosimeter placed on the anthropomorphic phantom's eye surface. The task transfer function (TTF), contrast-to-noise ratio (CNR), and low-contrast object specific CNR obtained from low-contrast phantom images reconstructed with filtered back projection (FBP), hybrid iterative reconstruction (HIR), and two types of DLR (DLRCTA and DLRLCD) were compared. In result, OB-TCM achieved a 32.5% dose reduction in the eye lens. Although HIR, DLRCTA, and DLRLCD showed lower TTF than FBP, the difference in TTF at the highest contributing spatial frequency corresponding to the contrast rod diameter was < 10%. Despite the OB-TCM acquisition, DLRCTA and DLRLCD achieved significantly lower noise and a higher CNR than FBP without OB-TCM (p < 0.05). However, low-contrast object specific CNR was equivalent among all reconstruction methods for the objective diameter of 5mm and slightly improved in DLRLCD for the objective diameter of 7mm. DLR with OB-TCM acquisition enabled dose reduction for the eye lens and high CNR image appearance, whereas the low contrast detectability evaluated by low-contrast object specific CNR did not always improve.

Dose-efficiency quantification of computed tomography systems using a model-observer.

Recent advances in computed tomography (CT) technology have considerably improved the quality of CT images and reduced radiation exposure in patients. At present, however, there is no generally accepted figure of merit (FOM) for comparing the dose efficiencies of CT systems. (i) To establish an FOM that characterizes the quality of CT images in relation to the radiation dose by means of a mathematical model observer and (ii) to evaluate the new FOM on different CT systems and image reconstruction algorithms. Images of a homogeneous phantom with four low-contrast inserts were acquired using three different CT systems at three dose levels and a representative protocol for CT imaging of low-contrast objects in the abdomen. The images were reconstructed using filtered-back projection and iterative algorithms. A channelized hotelling observer with difference-of-Gaussian channels was applied to compute the detectability ( ). This was done for each insert and each of the considered imaging conditions from square regions of interest (ROIs) that were (semi-)automatically centered on the inserts. The estimated detectabilities ( ) were averaged in the first step over the three dose levels ( ), and subsequently over the four contrast inserts ( ). All calculation steps included a dedicated assessment of the related uncertainties following accepted metrological guidelines. The determined detectabilities ( ) varied considerably with the contrast and diameter of the four inserts, as well as with the radiation doses and reconstruction algorithms used for image generation ( =1.3-5.5). Thus, the specification of a single detectability as an FOM is not well suited for comprehensively characterizing the dose efficiency of a CT system. A more comprehensive and robust characterization was provided by the averaged detectabilities and, in particular, . Our analysis reveals that the model observer analysis is very sensitive to the exact position of the ROIs. The presented automatable software approach yielded with the weighted detectability an objective FOM to benchmark different CT systems and reconstruction algorithms in a robust and reliable manner. An essential advantage of the proposed model-observer approach is that uncertainties in the FOM can be provided, which is an indispensable prerequisite for type testing.

ConvNeXt-based anchor-free object detection model for infrared image of power equipment

Infrared thermography has been widely used in the monitoring and diagnosis of power equipment to ensure the reliable functioning of power systems because of its non-contact and rapid capabilities. With the promotion of intelligent detection schemes in daily patrol inspection of power equipment by utilities, the amount of infrared images of power equipment (IIOPE) is growing geometrically. The research community has begun using deep learning technology to automatically detect objects in the IIOPE. However, the performance of existing detection methods is still not satisfactory because of the inherent attributes of power equipment infrared images, such as low contrast, complex background, and multi-scale object distribution. This research proposes an anchor-free detection model based on ConvNeXt to address these issues. An enhanced feature pyramid network (FPN) is presented to address the issue of multi-scale object distribution in the IIOPE. In addition, a dynamic soft label assignment strategy is proposed for precise object localization. Extensive experimental results on the self-built IIOPE dataset and the PASCAL VOC 2007 dataset demonstrate that the proposed method yields positive results. When compared to the current anchor-based as well as anchor-free infrared image detectors for power equipment, the proposed model offers improved accuracy while ensuring the detection speed that may satisfy the practical requirements.

CATER: Combined Animal Tracking &amp; Environment Reconstruction

Quantifying the behavior of small animals traversing long distances in complex environments is one of the most difficult tracking scenarios for computer vision. Tiny and low-contrast foreground objects have to be localized in cluttered and dynamic scenes as well as trajectories compensated for camera motion and drift in multiple lengthy recordings. We introduce CATER, a novel methodology combining an unsupervised probabilistic detection mechanism with a globally optimized environment reconstruction pipeline enabling precision behavioral quantification in natural environments. Implemented as an easy to use and highly parallelized tool, we show its application to recover fine-scale motion trajectories, registered to a high-resolution image mosaic reconstruction, of naturally foraging desert ants from unconstrained field recordings. By bridging the gap between laboratory and field experiments, we gain previously unknown insights into ant navigation with respect to motivational states, previous experience, and current environments and provide an appearance-agnostic method applicable to study the behavior of a wide range of terrestrial species under realistic conditions.