Small target detection using center-surround difference with locally adaptive threshold

Automatic target recognition (ATR) is a challenging task for several computer vision applications. It requires efficient, accurate, and robust methods for target detection and target identification. Deep learning has shown great success in many computer vision applications involving color RGB images. However, the performance of these networks in ATR with infrared sensor data needs further investigation. In this paper, we propose a multistage automatic target detection and recognition (ATDR) system that performs both target detection and target classification on infrared (IR) imagery using deep learning. Our system processes large IR image frames where targets take <1 % of the total number of pixels. First, we train a state-of-the-art object detector you only look once (YOLO) to localize all potential targets in the input image frame. Then, we train a convolutional neural network (CNN) to identify these detections as targets or false alarms. In this second phase, we adapt and analyze the performance of three CNN architectures: a compact and fully connected CNN, VGG16 with batch normalization, and a wide residual neural network (WRN). We also explore the use of a loss function that optimizes directly the area under the receiver operating characteristic (ROC) curve (AUC), and adapt it to our ATR application. To enhance the robustness of the proposed ATR to perturbation and variations introduced during the detection stage, we train our CNN classifiers on automatically detected targets using YOLO, in addition to ground truth bounding boxes and apply selected data augmentation techniques. To simulate real testing environments, where the spatial location of the targets within the image frame is unknown, only YOLO-detected boxes are used during validation. We evaluate our ATDR on a real benchmark dataset that includes different vehicles captured at different resolutions. Our experiments have shown that YOLO can detect most of the targets at the expense of generating a high number of false alarms. We show that the VGG-16 network with batch normalization, which is the best performing model, can correctly identify the classes of the targets, as well as classify the majority of YOLO’s false detections into an additional nontarget class. We also show that the proposed training modification to optimize an AUC-based loss function for ATR proved to be advantageous mainly in identifying difficult targets.

Automatic extraction of salient regions is beneficial for various computer vision applications, such as image segmentation and object recognition. The salient visual information across images is very useful and plays a significant role for the visually impaired in identifying tactile information. In this paper, we introduce a novel saliency cuts method using local adaptive thresholding to obtain four regions from a given saliency map. First, we produced four regions for image segmentation using a saliency map as an input image and local adaptive thresholding. Second, the four regions were used to initialize an iterative version of the GrabCuts algorithm and to produce a robust and high-quality binary mask with a full resolution. Finally, salient objects’ outer boundaries and inner edges were detected using the solution from our previous research. Experimental results showed that local adaptive thresholding using integral images can produce a more robust binary mask compared to the results from previous works that make use of global thresholding techniques for salient object segmentation. The proposed method can extract salient objects with a low-quality saliency map, achieving a promising performance compared to existing methods. The proposed method has advantages in extracting salient objects and generating simple, important edges from natural scene images efficiently for delivering visually salient information to the visually impaired.

Automatic target recognition (ATR) is an important part for many computer vision applications. Despite the extensive research which has been carried out in this area for many years, there is no ATR system which performs well on all applications. Recently, different object recognition frameworks have been proposed which yield a high performance in baseline databases. However, our experiments showed that they can fail in real-world scenarios, when dealing with a limited number of data samples. In this paper, we propose a new ATR system, based on deep convolutional neural network (DCNN), to detect the targets in forward looking infrared (FLIR) scenes and recognize their classes. In our proposed ATR framework, a fully convolutional network is trained to map the input FLIR imagery data to a fixed stride correspondingly-sized target score map. The potential targets are identified by applying a threshold on the target score map. Finally, the corresponding regions centered at these target points are fed to a DCNN to classify them into different target types while at the same time rejecting the false alarms. The proposed architecture achieves a significantly better performance in comparison with that of the state-of-the-art methods on two large FLIR image databases.

Forward Looking Infrared (FLIR) automatic target recognition (ATR) systems depend upon the capacity of the atmosphere to propagate thermal radiation over long distances. To date, not much research has been conducted on analyzing and mitigating the effects of the atmosphere on FLIR ATR performance, even though the atmosphere is often the limiting factor in long-range target detection and recognition. The atmosphere can also cause frame-to-frame inconsistencies in the scene, affecting the ability to detect and track moving targets. When image quality is limited by turbulence, increasing the aperture size or improving the focal plane array cannot improve ATR performance. Traditional single frame image enhancement does not solve the problem. A new approach is described for reducing the effects of turbulence. It is implemented under a lucky-region-imaging framework using short integration time and spatial domain processing. It is designed to preserve important target and scene structure. Unlike previous Fourier-based approaches originating from the astronomical community, this new approach is intended for real-time processing from a moving platform, with ground as the background. The system produces a video stream with minimal delay. A new video quality measure (VQM_turb) is presented for quantifying the success of turbulence mitigation on real data where no reference imagery is available. The VQM_turb is the core of the innovation because it allows a wide range of algorithms to be quantitatively compared. An algorithm can be chosen, and then tuned, to best-fit available processing power, latency requirements, scenarios and sensor characteristics.

In the field of infrared remote sensing, the problem of IR small target detection is still an important component part. Concerning infrared dim small target (IRDST) detection, firstly the infrared image is processed with DWT method to get the wavelet coefficients image, but the distribution characteristics, such as scales, frequencies, orientations of wavelet coefficients in different sub-bands are various, so wavelet image denoised by a single threshold criterion can not give a satisfying estimation. Based on this motivation, a spatially adaptive multi-model de-noising strategy (SAMMDS) based IRDST detection method is proposed in this paper, which can adjust thresholding strategy according to the distribution of noise in different scales and directions. Spatially adaptive BayesShrink (SABS) thresholding, traditional BayesShrink (BS) thresholding and generalized cross validation (GCV) thresholding are all adopted here to process each sub-band separately. After reconstructing the denoised wavelet image, a simple global thresholding is used to separate the background and target finally. Experimental results demonstrate that the proposed algorithm performs better than other typical wavelet methods for small target detection with various complex backgrounds.

Small and dim target detection is a critical challenge in hyperspectral remote sensing, particularly in complex, large-scale scenes where spectral variability across diverse land cover types complicates the detection process. In this paper, we propose a novel target reasoning algorithm named Adaptive Global Dense Nested Reasoning Network (AGDNR). This algorithm integrates spatial, spectral, and domain knowledge to enhance the detection accuracy of small and dim targets in large-scale environments and simultaneously enables reasoning about target categories. The proposed method involves three key innovations. Firstly, we develop a high-dimensional, multi-layer nested U-Net that facilitates cross-layer feature transfer, preserving high-level features of small and dim targets throughout the network. Secondly, we present a novel approach for computing physicochemical parameters, which enhances the spectral characteristics of targets while minimizing environmental interference. Thirdly, we construct a geographic knowledge graph that incorporates both target and environmental information, enabling global target reasoning and more effective detection of small targets across large-scale scenes. Experimental results on three challenging datasets show that our method outperforms state-of-the-art approaches in detection accuracy and achieves successful classification of different small targets. Consequently, the proposed method offers a robust solution for the precise detection of hyperspectral small targets in large-scale scenarios.

In recent years, there has been an increased scope for assistive software and technologies, which help the visually impaired to perceive and recognize natural scene images. In this article, we propose a novel saliency cuts approach using local adaptive thresholding to obtain four regions from a given saliency map. The saliency cuts approach is an effective tool for salient object detection. First, we produce four regions for image segmentation using a saliency map as an input image and applying an automatic threshold operation. Second, the four regions are used to initialize an iterative version of the Grab Cut algorithm and to produce a robust and high-quality binary mask with a full resolution. Lastly, based on the binary mask and extracted salient object, outer boundaries and internal edges are detected by Canny edge detection method. Extensive experiments demonstrate that the proposed method correctly detects and extracts the main contents of the image sequences for delivering visually salient information to the visually impaired people compared to the results of existing salient object segmentation algorithms

Thresholding is a simple and effective technique for image segmentation. Thresholding techniques can be grouped into two categories, global thresholding and local thresholding. All local threshold method generally begins with determining thresholds in each pixel by checking the area centered on the pixel, using a box shape (x, y) which is fixed by the size of the neighborhood b. If the neighborhood is very small, then the algorithm will be sensitive to noise and excessive segmentation occurs. Whereas, if the size of the neighborhood is very large then the algorithm will apply resemble the global threshold method. In this study, we propose a method of calculation of Local Adaptive Region, to determine the value of each pixel that is flexible neighborhoods, where each pixel has values different neighborhoods based on the value of the standard deviation region. Adaptive method on the local region thresholding consists of several processes, namely: Image Enhancement, Adaptive Local Region and thresholding. Based on evaluation of ME, image result of threshold using the Adaptive Local Region method, giving an average ME smallest value, that is 16.99% at Niblack method and 19.46% at Sauvola method. And on evaluation of the RAE, image result of threshold using the Adaptive Local Region method, giving an average RAE smallest value, that is 15.26% at Niblack method and 25.58% at Sauvola method. In addition, the results of trials with various noise variance represent that the method of Adaptive Local Region resistant to noise.

Kanker serviks merupakan ancaman kesehatan global serius, dengan WHO melaporkan sekitar 604.000 kasus baru dan 342.000 kematian pada tahun 2020. Penelitian ini mengeksplorasi kombinasi metode local adaptive threshold dan segmentasi watershed untuk meningkatkan akurasi deteksi dini kanker serviks dengan lebih akurat mengidentifikasi sel-sel yang saling tumpang tindih pada Pap Smear. Metode Local Adaptive Threshold menyesuaikan nilai ambang berdasarkan karakteristik lokal gambar, dan segmentasi watershed diaplikasikan untuk memisahkan sel-sel yang saling tumpang tindih. Kombinasi ini menunjukkan hasil yang menjanjikan dalam meningkatkan efisiensi dan akurasi skrining kanker serviks, mendukung strategi WHO untuk eliminasi kanker serviks. Namun, adopsinya menghadapi tantangan di negara berkembang karena keterbatasan sumber daya dan kesenjangan digital. Tes menggunakan K-Fold Cross Validation (5 dan 7) menunjukkan akurasi 90.93% untuk k=5, dengan rata-rata precision 97.97%, recall 49.22%, dan F1-Score 65.50%. Pada k=7, hasil sedikit meningkat dengan precision 97.99%, recall 49.24%, dan F1-Score 65.53%. Rata-rata PSNR adalah 43.4341 dB dan MSE 3.45061, menegaskan efektivitas metode.Kata Kunci: Local Adaptive Thresholding, Watershed, Cervical Cancer, Pap Smear

Speech recognition of continuous speech will be greatly influenced by the word segmentation process of the speech input. Proper segmentation will result in better speech recognition. This study proposed automatic speech recognition by applying local adaptive thresholding in the segmentation process. The segmentation method used is an enhanced blocking block area method whose input is a spectrogram image of the speech signal. While the locally adaptive thresholding method used is the Niblack method which is the best method compared to other methods, namely Sauvola, Bradley, Guanglei Xiong, and Bernsen when applied to the enhanced blocking block area method. For the speech recognition process, using Mel-frequency cepstral coefficients (MFCC) as a feature extraction method and Hidden Markov Model (HMM) as speech recognition. The experimental results show that by using 400 sentences consisting of 80 testing data and 320 as training data and using K-fold cross-validation, the highest accuracy is 60,8%. This result has no significant difference with the use of global thresholding in the segmentation process.

Combining the capability of an automated target recognition (ATR) system with the expertise of human operators has the potential to improve combined human-machine target-recognition performance. The objective of this work was to develop empirically validated guidelines for the display of information provided by an ATR system. Motivated by the premise that operators will have higher trust in an ATR system that justifies its decisions, this work explored alternative ATR displays that provided different types and amounts of information in support of the system's target decisions. Operators' usage of and preference for the displays were contrasted for low- and high-accuracy simulated ATR systems representative, respectively, of current and future technologies. Performance results suggested that the amount of supporting information that is displayed must be congruent with the ATR system's accuracy. Operators expressed a clear preference for an ATR system that provides a confidence rating for its target decisions. Based on the results design guidelines are suggested for displaying ATR information to help operators balance confidence in their own judgments with trust in the ATR's decisions.

Presents an implementation of forward-looking infrared (FLIR) and laser radar (LADAR) data simulation for use in developing a multi-sensor data-fusion automated target recognition (ATR) system. Through the use of commercial models and software, we can create a highly detailed scene model, which provides a rich data set for later processing. We embed our own modules within this software to extract data from the scene and then present it to the FLIR and LADAR sensor models. These models produce simulated range and temperature readings for objects within the scene. These data are used to create a variety of images to aid in the visualisation of the data and to test our ATR system. Frames from approach sequences show the LADAR and FLIR data in different formats. The simulated data and subsequent images are accurate and rapid to produce, and provide invaluable data resources for testing an ATR system.

Research on Infrared Dim and Small Target Detection Algorithm Based on Low-Rank Tensor Recovery

&lt;p&gt;Continuous speech is a form of natural human speech that is continuous without a clear boundary between words. In continuous speech recognition, a segmentation process is needed to cut the sentence at the boundary of each word. Segmentation becomes an important step because a speech can be recognized from the word segments produced by this process. The segmentation process in this study was carried out using local adaptive thresholding technique in the blocking block area method. This study aims to conduct performance comparisons for five local adaptive thresholding methods (Niblack, Sauvola, Bradley, Guanglei Xiong and Bernsen) in continuous speech segmentation to obtain the best method and optimum parameter values. Based on the results of the study, Niblack method is concluded as the best method for continuous speech segmentation in Indonesian language with the accuracy value of 95%, and the optimum parameter values for such method are window = 75 and k = 0.2.&lt;/p&gt;

Stripe noise is usually introduced into infrared images during acquisition, which seriously deteriorates the image quality and affects the subsequent analysis of the image. In this paper, the spatial mathematic model and the comb-like impulse spectrum characteristics in Fourier domain of stripe noise are given. On this basis, a local adaptive threshold noise detection linear interpolation filter (LALIF) for stripe noise removal in infrared images is designed. In the filter method, the first, we crop the original image into two sub-images with optimal size to make the comb-like impulse spectrum characteristics of the stripe noise most prominent. The second, we use a local adaptive threshold to distinguish noise frequency points from the useful frequency points in the peak regions of the amplitude spectrum, and reassign the value of noise frequency points by linear interpolation. At last, we amalgamate the sub-images to obtain the stripe noise removal image. Experimental results indicate that the proposed method has a good performance in adaptability and stability for complex stripe noise removal in infrared images.