Search

Computer vision is nowadays is one of the promising & evolving areas in information technology. Image processing is increasingly used in several applications such as automotive, medical or aerospace. Every computer vision application involves capturing several images with the help of camera. Image acquisition, image transmission & image security are the three important aspects of image processing system. This paper includes development of complete image processing system including improved median filtering with better PSNR & operating frequency, Image compression module with better PSNR & compression ratio, Image Encryption module using Advanced Encryption Standard.

Recent transformer-based models, especially patch-based methods, have shown huge potentiality in vision tasks. However, the split fixed-size patches divide the input features into the same size patches, which ignores the fact that vision elements are often various and thus may destroy the semantic information. Also, the vanilla patch-based transformer cannot guarantee the information communication between patches, which will prevent the extraction of attention information with a global view. To circumvent those problems, we propose the Efficient Attention Pyramid Transformer (EAPT) in this paper. More specifically, we first propose Deformable Attention, which learns an offset for each position in patches. Therefore, even with split fixed-size patches, our method can still obtain non-fixed attention information that can cover various vision elements. Then, we design the Encode-Decode Communication module (En-DeC module), which can obtain communication information between all patches to get more complete global attention information. Finally, we also proposed a position encoding specifically for vision transformers, which can be used for patches of any dimensions and any lengths. Extensive experiments on the vision tasks of image classification, object detection, semantic segmentation demonstrate the effectiveness of our proposed model. Furthermore, we also conduct rigorous ablation studies to evaluate the key components of the proposed structure.

Human-machine interactive systems show increasing demand for analysing fast moving objects in high-frame-rate videos. Robust foreground detection, which is able to reduce large amount of redundant background data from high-frame-rate video, becomes the essence to achieve ultra-high-speed human-machine interactions. This paper proposes a local spatial propagation based background model generation, a local linear illumination correction based background model update, and a regional central coordinates and edge keypoints constrained foreground region reselection. The three proposals make up a robust and hardware-friendly foreground detection method. Experimental results prove that the proposed hardware-friendly algorithm achieves high accuracy and robustness on various kinds of challenging cases. Meanwhile, the hardware implementation utilizes little hardware resources and achieves realtime processing of high-frame-rate (784 frame/second) video with the delay less than 1 ms/frame in image processing core. In addition, a practical system is implemented by combing a PC, a high-speed camera and a field programmable gate array (FPGA) for realworld applications. This work will significatively promote the development and application of high-speed human machine interaction. A demo of the proposed vision system working at 784 FPS is available at https://wcms.waseda.jp/em/5f84f75136a6.

In robot visual servoing system, effective detected the region of interest area (ROI) in target images is the first problem should be solved, but the detection is susceptible to an unstructured environment. In this paper, an instance segmentation network algorithm based on Mask Region Convolutional Neural Networks (Mask R-CNN) framework is proposed for ROI image preprocessing. As instance segmentation technology can distinguish complex environmental information, so first with this advantage filter out message such as shape or similar area to target image, then with semantic segmentation technology add special category labels to the filtered images and distinguish different individual instances in similar categories. Finally through the series steps aforementioned, robot vision system can overcome the impact of environmental factors and identify the target image. The proposed method is used to detect target image under five constraints such as occlusion and reflection, result shows the algorithm can effectively deal with challenges that brought by complex constraints, and even can predict the location data of missing information based on some image information. In addition, based on the algorithm proposed in this paper, we used one seven axis robot visual servo platform, executed visual servoing experiment under different unstructured environments, further verifies the effectiveness of our proposed method.

The data processing of iris detection which was done by computer system were able to get better and eficient information. One of method that was used in digital image processing is edge detection. The function of this method is to identify the line or edge of the image so that the information of border line of image that can be highlighted and applied to identify the iris digital image pattern. In this research, the researcher has been tried an alternative to segment the system in order to identify the iris focused to image processing by edge detection and iris fitur extraction used Sobel detection. Iris image was from Chinese Academy of Science Institute of Automation (CASIA) version 1.0 database. The images were gray level images in BMP format and 320 x 280 pixel. It was consists of 756 images. As the results, 20 selected iris digital images randomly could not be identified by Sobel detection.

With the growing demand for geospatial data, challenging aerial images with high spatial, spectral, and temporal resolution achieve excellent development. Currently, deep Convolutional Neural Network (CNN) structures are applied widely for object detection. Nevertheless, existing deep CNN-based models consist of complex network structures and require immense amounts of graphics processing unit (GPU) computation power with high energy consumption. Thus, achieving efficient real-time object detection for limited memory and processing capacity embedded device is a major challenge. This paper proposes a feasible and lightweight object detection model based on deep CNN where a mobile inverted bottleneck module is adopted in the backbone structure. Moreover, an enhanced spatial pyramid pooling is adopted to increase the receptive field in the network by concatenating the multi-scale local region features. The experimental results demonstrated that the proposed model achieved higher average precision and required the smallest memory storage compared to previous works. Moreover, the proposed model offers the best trade-offs in terms of detection accuracy, model size, and detection time which has excellent potential to be deployed on limited capacity embedded device.

The speckle noise is the major area of concern in ultrasound imaging. Speckle degrades the speed and accuracy of subsequent image processing tasks such as segmentation, description etc. But reduction of speckle may cause blurring or loss of edges and important features. In our work, we have employed a novel thresholding rule based on wavelet transform for speckle reduction from ultrasound images. The wavelet transform performs multi-scale analysis of the given image by treating different frequency components present in an image separately. The experiment results exhibit that the proposed thresholding rule gives better results for speckle reduction, edge preservation and feature preservation for medical ultrasound images, as compared with the existing thesholding rules.

— The use of image recognition technology has become increasingly popular in recent years, with applications ranging from social media platforms to e-commerce websites. In this project, we uses a Convolutional Neural Network (CNN) algorithm for image recognition. Convolutional Neural Networks (CNNs) are a type of deep neural network that have revolutionized the field of computer vision and image recognition. With their ability to learn complex patterns and features from images, CNNs have become the backbone of many modern computer vision applications.

— Using OpenCV and PyZbar, this session exposes attendees to the realm of Bar code scanning. Participants will investigate how real-time Bar code scanning applications can be used in practise. The course introduces the fundamentals of video capture with OpenCV and shows how to recognise and decode Bar codes with PyZbar. Also covered are methods for improving the user interface by superimposing visual cues over discovered Bar codes. Participants will learn how to combine scanning codes with databases so that actions based on decoded data are possible. The use of session variables to hold user information is highlighted in the webinar.

Background/Purpose: These days, the involvement of computer science in agriculture and food science is expanding. Classification and fault identification of diverse products employ a variety of Artificial Intelligence (AI), soft computing approaches, and methodologies, which contribute to higher-quality products for consumers. The position of Arecanuts in the international and Indian markets, as well as the application of computer vision and image processing to a system for categorizing and grading Arecanuts, are the main topics of this article. Objective: The development of a system for the automated categorization of Arecanut using images is limited by difficulties. To assess the value of computer vision application for Arecanut, it is critical to taken as account the traditional and economic significance of Arecanut. Design/Methodology/Approach: Several types of Arecanut are prone to great variation in color, texture, and form depending on the category and the area in which they are cultivated. Arecanuts are processed utilizing a variety of techniques, with an emphasis on the finished product's exterior. Here, the color, size, and texture of Arecanut are used to construct a classification or grading system. Findings/Result: With reference to the cited significant work that has been done on other fruits as well as Arecanuts from the standpoint of computer vision. This article provided a thorough introduction to Arecanuts, computer vision, and the uses and benefits of vision-aided technologies in the grading of Arecanuts and categorization. Result Limitations/Implications: This review is based on the detection and classification of the Arecanuts done using computer vision and AI techniques. Originality Value: Several inline resources including review papers on Arecanut, research articles, technical books, and website resources. Paper Type: Literature Review paper on smart auto Arecanut Sorting and Grading of Arecanut using Computer Vision and Image Processing

RGB-D cameras are a type of imaging device that includes a standard colour camera and a depth sensor. The depth sensor records depth images, where each pixel measures the distance of the target from the camera. Using data from the RGB-D camera, we can recon- struct a three-dimensional scene. Since the introduction of the Microsoft Kinect, multiple affordable RGB-D cameras have reached the market and are even integrated into laptops and mobile phones. In this thesis, healthcare applications of RGB-D cameras are investigated, focussing on two areas. The first is Structured Light Plethysmography (SLP), a method for measuring lung function. The non-contact nature of SLP is ideal for respiratory measurement, as it avoids potential contamination without requiring single-use equipment. The current SLP device is large and expensive, while the data it produces is low-resolution and does not adapt to the patient. A new SLP processing pipeline based on an RGB-D camera is proposed, which will produce automatically aligned, high-resolution data in a portable device. Algorithm development to enhance the data processing capabilities of this new pipeline is presented. A method for interpreting data from the high-resolution grid, via a control point-based shape modelling technique, is used to show differences in the respiratory pat- tern before and after exercise. A system is described for compensating for the motion of the subject, so the resulting data only shows motion due to breathing. This method is demonstrated by compensating for the pedalling motion of a subject on an exercise bike. The second application area is the Neonatal Intensive Care Unit (NICU), which cares for critically ill new-born babies. A proof-of-concept study is designed and conducted to investigate the potential use of an RGB-D camera in the NICU. In this thesis, the collec- tion of 24 hours of RGB-D video footage is presented, and it is shown how the infant’s respiratory rate can be measured from the depth data.

Image-based plant phenotyping represents a transdisciplinary research domain that utilises modern non-invasive technology combined with time-series high-throughput measurements for the extensive analysis and evaluation of plant traits such as morphology, physiology, and biochemistry. Phenotyping can be applied to an individual or group of experimental plants in various environments such as a laboratory, open or closed growth chamber, glasshouse, or field. Recognition of plant accessions plays an important role while evaluating, selecting and producing cultivars. However, it is very challenging, time-consuming and near impossible to correctly identify multiple accessions purely based on visual inspection or through manual phenotypic measurements. This research focuses on recognising different Arabidopsis accessions using high-resolution RGB images collected in two independent experiments with 40 Arabidopsis accessions and a high number of replicates. Experiments are conducted in an automated indoor plant cultivation chamber composed of full temperature control and LED sources ranging from 400 to 750 nm. Following data acquisition, various deep learning models are assessed to achieve interpretable accuracy of accession recognition by using a single image or a group of images taken over several days. Performance of selected models is scrutinised by classifying accessions while receiving a single image at the input or in a second case series of images whose number depends on the pre-defined screening days (for example, 3, 7, or 12 days after screening). Owing to the high variability among replicates in the majority of accessions, data complexity has to be reduced by extracting general leaf phenotypic traits and re-defining data split using statistical methods such as similarity matrix. This leads to achieving the desired accuracy (>92%) and precision in terms of correct recognition of various Arabidopsis accessions. The best-performing deep learning models utilised in this computer vision task could be further applied in the glasshouse environment to identify unknown accessions and therefore predict plant growth. https://icar2023.org/program/

Indiana University-Purdue University Indianapolis (IUPUI), In traffic scene perception for autonomous vehicles, driving videos are available from in-car sensors such as camera and LiDAR for road detection and collision avoidance. There are some existing challenges in computer vision tasks for video processing, including object detection and tracking, semantic segmentation, etc. First, due to that consecutive video frames have a large data redundancy, traditional spatial-to-temporal approach inherently demands huge computational resource. Second, in many real-time scenarios, targets move continuously in the view as data streamed in. To achieve prompt response with minimum latency, an online model to process the streaming data in shift-mode is necessary. Third, in addition to shape-based recognition in spatial space, motion detection also replies on the inherent temporal continuity in videos. While current works either lack long-term memory for reference or consume a huge amount of computation. The purpose of this work is to achieve strongly temporal-associated sensing results in real-time with minimum memory, which is continually embedded to a pragmatic framework for speed and path planning. It takes a temporal-to-spatial approach to cope with fast moving vehicles in autonomous navigation. It utilizes compact road profiles (RP) and motion profiles (MP) to identify path regions and dynamic objects, which drastically reduces video data to a lower dimension and increases sensing rate. Specifically, we sample one-pixel line at each video frame, the temporal congregation of lines from consecutive frames forms a road profile image; while motion profile consists of the average lines by sampling one-belt pixels at each frame. By applying the dense temporal resolution to compensate the sparse spatial resolution, this method reduces 3D streaming data into 2D image layout. Based on RP and MP under various weather conditions, there have three main tasks being conducted to contribute the knowledge domain in perception and planning for autonomous driving. The first application is semantic segmentation of temporal-to-spatial streaming scenes, including recognition of road and roadside, driving events, objects in static or motion. Since the main vision sensing tasks for autonomous driving are identifying road area to follow and locating traffic to avoid collision, this work tackles this problem by using semantic segmentation upon road and motion profiles. Though one-pixel line may not contain sufficient spatial information of road and objects, the consecutive collection of lines as a temporal-spatial image provides intrinsic spatial layout because of the continuous observation and smooth vehicle motion. Moreover, by capturing the trajectory of pedestrians upon their moving legs in motion profile, we can robustly distinguish pedestrian in motion against smooth background. The experimental results of streaming data collected from various sensors including camera and LiDAR demonstrate that, in the reduced temporal-to-spatial space, an effective recognition of driving scene can be learned through Semantic Segmentation. The second contribution of this work is that it accommodates standard semantic segmentation to sequential semantic segmentation network (SE3), which is implemented as a new benchmark for image and video segmentation. As most state-of-the-art methods are greedy for accuracy by designing complex structures at expense of memory use, which makes trained models heavily depend on GPUs and thus not applicable to real-time inference. Without accuracy loss, this work enables image segmentation at the minimum memory. Specifically, instead of predicting for image patch, SE3 generates output along with line scanning. By pinpointing the memory associated with the input line at each neural layer in the network, it preserves the same receptive field as patch size but saved the computation in the overlapped regions during network shifting. Generally, SE3 applies to most of the current backbone models in image segmentation, and furthers the inference by fusing temporal information without increasing computation complexity for video semantic segmentation. Thus, it achieves 3D association over long-range while under the computation of 2D setting. This will facilitate inference of semantic segmentation on light-weighted devices. The third application is speed and path planning based on the sensing results from naturalistic driving videos. To avoid collision in a close range and navigate a vehicle in middle and far ranges, several RP/MPs are scanned continuously from different depths for vehicle path planning. The semantic segmentation of RP/MP is further extended to multi-depths for path and speed planning according to the sensed headway and lane position. We conduct experiments on profiles of different sensing depths and build up a smoothly planning framework according to their them. We also build an initial dataset of road and motion profiles with semantic labels from long HD driving videos. The dataset is published as additional contribution to the future work in computer vision and autonomous driving.

Due to the recent pandemic, electronic know-your-customer (e-KYC) technologies have enabled business continuity for various sectors. In the case of Malaysia, the Malaysia National Identity card (MyKad) is often used for identity verification purposes. However, when presenting MyKad online, the uploaded MyKad could be an image of the actual MyKad taken a few years back, a printout of a legit MyKad, or a digitally/physically tampered MyKad. This research aims to detect non-genuine MyKad. Specifically, we consider the unique colour count, the surface roughness score (i.e., complexity), and local binary patterns (LBP) of the received MyKad image to make inferences.

With the development of technology, the human face could be used as a key to accessing critical resources. Understanding the potentiality of the human face is crucial before using unique facial features in the face application. As such, the novel Local Directional Texture (LDT) and its variants, temporal strips (TS) and action unit temporal blocks (AU-TB), are proposed for privacy-preserving face recognition. The standard datasets are also analysed to revisit the action units (AUs) encoded in the dataset ground truths and inspect whether any significant AU has been left out. Lastly, a graph-based algorithm is developed to classify the micro-expressions.

Internet of things (IoT) becomes the backbone of the advanced countries and it has a real contribute to exchange the traditional style or way of practical life, even personal life into smart style, with (IoT) technology the life become more and more easy and professional. internet of things achieves various applications coordinate with sensors and standard protocols to apply what is called machine -to- machine connection (M2M), in this paper we will talk more about the concept of (M2M), the main component of internet of things and finally the common protocols that is used in network, in addition to that this work present an IOT operation with processing system using camera for capturing image and Xilinx system generator(XSG)models for designing image processing algorithms and the result of the processing is an image with black and white for edge detection and Thresholding models and gray color image for gray enhancement model.

Meeting the nutritional requirements of an increasing population in a changing climate is the foremost concern of agricultural research in recent years. A solution to some of the many questions posed by this existential threat is breeding crops that more efficiently produce food with respect to land and water use. A key aspect to this optimization is geometric aspects of plant physiology such as canopy architecture that, while based in the actual 3D structure of the organism, does not necessarily require such a representation to measure. Although deep learning is a powerful tool to answer phenotyping questions that do not require an explicit intermediate 3D representation, training a network traditionally requires a large number of hand-segmented ground truth images. To bypass the enormous time and expense of hand- labeling datasets, we utilized a procedural sorghum image pipeline from another student in our group that produces images similar enough to the ground truth images from the phenotyping facility that the network can be directly used on real data while training only on automatically generated data. The synthetic data was used to train a deep segmentation network to identify which pixels correspond to which leaves. The segmentations were then processed to find the number of leaves identified in each image to use for the leaf-counting task in high-throughput phenotyping. Overall, our method performs comparably with human annotation accuracy by correctly predicting within a 90% confidence interval of the true leaf count in 97% of images while being faster and cheaper. This helps to add another expensive- to-collect phenotypic trait to the list of those that can be automatically collected.

The performance of long-range imaging systems often suffers due to the presence of atmospheric turbulence. One way to alleviate the degradation caused by atmospheric turbulence is to apply post-processing mitigation algorithms, where a high-quality frame is reconstructed from a single degraded image or a sequence of degraded frames. The image processing algorithms for atmospheric turbulence mitigation have been studied for decades, yet some critical problems remain open. This dissertation addresses the problem of image reconstruction through atmospheric turbulence from three unique perspectives: 1) Reconstruction with the presence of moving objects using an improved classical image processing pipeline. 2) A fast simulation scheme for efficiently generating large-scale turbulence-degraded datasets for training deep neural networks. 3) A deep learning-based single-frame reconstruction method using Vision Transformer.

Imaging and computer vision systems deployed in real-world environments face the challenge of accommodating a wide range of lighting conditions. However, the cost, the demand for high resolution, and the miniaturization of imaging devices impose physical constraints on sensor design, limiting both the dynamic range and effective aperture size of each pixel. Consequently, conventional CMOS sensors fail to deliver satisfactory capture in high dynamic range scenes or under photon-limited conditions, thereby impacting the performance of downstream vision tasks. In this thesis, we address two key problems: 1) exploring the utilization of spatial multiplexing, specifically spatially varying exposure tiling, to extend sensor dynamic range and optimize scene capture, and 2) developing techniques to enhance the robustness of object detection systems under photon-limited conditions. In addition to challenges imposed by natural environments, real-world vision systems are susceptible to adversarial attacks in the form of artificially added digital content. Therefore, this thesis presents a comprehensive pipeline for constructing a robust and scalable system to counter such attacks.

Median filter is one of the predominant filters that are used to suppress impulse noise. Its simplicity and ability to maintain edges has led to an extensive application in the domain of image processing and computer vision. However, challenges such as moderate to high running time of the standard median filter algorithm and relatively poorer performance when the image is highly corrupted with impulse noise, have led to the design of several variations of the algorithm. One set of variation of the algorithm concentrates on generating quality outputs, while the other set focuses on reducing running time. Among the set targeting the reduction of the running time of the median filter is the DP approximated median filter. However, DP performs poorly when images are corrupted with moderate to high levels of noise. This paper therefore proposes an Improved Approximation Median Filtering Algorithms (IAMFA-I & IAMFA-II) based on DP to generate a better output. The introduction of Mid-Value-Decision-Median in DP reduces the chances of selecting corrupted pixel for denoised image. Experimental results indicate that the IAMFA-II has better running time and equivalent output compared with DP, while IAMFA-I generates better output and has equivalent running time when compared with DP.

Synthetic aperture radar (SAR), due to its inherent characteristics, will produce speckle noise, which results in the deterioration of image quality, so the removal of speckle in SAR image is very important for the subsequent high-level image processing. In order to balance the relationship between denoising and texture preservation, we propose a multiscale residual dense dual attention network (MRDDANet) for SAR image denoising. This algorithm can effectively suppress the speckle while fully retaining the texture details of the image. In MRDDANet, shallow features are extracted from the noisy images by multiscale modules with different kernel sizes, and then, the extracted shallow features are mapped to the residual dense dual-attention network to obtain the deep features of SAR image. Finally, the final denoising image is generated through global residual learning. MRDDANet has advantages of both multiscale blocks and residual dense dual attention networks. The dense connection can fully extract features in the image, and the dual-channel attention enables MRDDANet to pay more attention to noise information, which is beneficial to remove noise and keep the details of the original image at the same time. Compared with state-of-the-art algorithms, the results of the experiment indicate that our method not only improves various objective indicators but also shows great advantages in visual effects.

The structural dynamics of the amyloid protein aggregation process are associated with neurodegenerative diseases, including Alzheimer's disease and Parkinson's disease. High-speed atomic force microscopy (HS-AFM) is able to visualize the structural dynamics of individual aggregate species that otherwise cannot be distinguished. HS-AFM observations also detect impurities in the sample, and thus, experiments require relatively high sample purity. To derive valid information regarding the structural dynamics of the sample from the high-speed AFM images, a correction of the influence caused by the drift of the stage (scanner) from all frames is required. However, correcting the HS-AFM videos that consist of a large number of images requires significant effort. Here, using HS-AFM observation of α-synuclein fibril elongation as an example, we propose an HS-AFM image processing procedure to correct stage drift in the x-, y-, and z-directions with the free software ImageJ. ImageJ with default settings and our plugins attached to this article can process and analyze image stacks, which allow users to easily detect and show the temporal change in sample structures. This processing method can be automatically applied to numerous HS-AFM videos by batch processing with a series of ImageJ macrofunctions.

In the field of remote sensing image processing, clouds heavily affect the quality of the remote sensing images and their application potential. Thus, in recent years, with the prevalence of deep learning techniques used in the field of image processing, many methods have been proposed for cloud removal using single remote sensing images. The existing single-image cloud removal methods suffer from poor generalization capabilities that prevent them from being applied to diverse remote sensing images. Thus, a novel method using a multimodal architecture is proposed which provides multiple most likely outputs for the image and selects the best one through perception-based image quality evaluator (PIQE). In addition, adversarial consistency loss is used to replace cycle consistency loss, which encourages the model to retain more texture information of the original image, and thus the quality of the generated image increases. Experiments demonstrate that the presented method can easily achieve a considerable increase in the peak signal-to-noise ratio and the structural similarity index compared with other methods.

In today’s era, there is a huge rise in demand for the data and information available in a printed medium. To resume time, cost, and energy, we have been developing numerous technology. One of them, convert JPG, JPEG, or PDF softcopy into text, this process calls Text extraction. Identification of characters and detection of words and texts is again an upfront challenging task. This paper is focused on an investigation of feature and text extraction from images using image recognition in Android. In this era of Research and Technology, Image processing has become one of the most critical fields. Using a number of researchers' papers, we are going to study appropriate methods or algorithm, their accuracy, a key feature of each paper. All research papers going to study in the context of Android technology.

PURPOSE Involuntary patient movement results in data discontinuities during computed tomography (CT) scans which lead to a serious degradation in the image quality. In this paper, we specifically address artifacts induced by patient motion during a head scan. METHOD Instead of trying to solve an inverse problem, we developed a motion simulation algorithm to synthesize images with motion induced artifacts. The artifacts induced by rotation, translation, oscillation and any possible combination are considered. Taking advantage of the powerful learning ability of neural network, we designed a novel 3D network structure with both a large reception field and a high image resolution to map the artifact free images from artifact contaminated images. Quantitative results of the proposed method were evaluated against the results of U-Net and proposed networks without dilation structure. 30 sets of motion contaminated images from two hospitals were selected to do clinical evaluation. RESULT Facilitating the training dataset with artifacts induced by variable motion patterns and the neural network, the artifact can be removed with good performance. Validation dataset with simulated random motion pattern showed outperformed image correction, and quantitative results showed the proposed network had the lowest normalized root mean-square-error, highest peak signal-to-noise ratio and structure similarity, indicating our network gave the best approximation of gold standard. Clinical image processing results further confirmed the effectiveness of our method. CONCLUSION We proposed a novel deep learning based algorithm to eliminate motion artifacts. The CNNs trained with synthesized image pairs achieved promising result in artifacts reduction. The corrected images increased the diagnostic confidence compared with artifacts contaminated images. We believe that the correction method can restore the ability to successfully diagnose and avoid repeated CT scans in certain clinical circumstances. This article is protected by copyright. All rights reserved.

A method for processing microfocus X-ray images is described. It is based on high-frequency filtration and morphological image processing, which increases the contrast of the X-ray details. One of the most informative X-ray techniques is microfocus X-ray. In some cases, microfocus X-ray images cannot be reliably analyzed due to the peculiarities of the shooting method. So, the main disadvantages of microfocus X-ray images are most often an uneven background, distorted brightness characteristics and the presence of noise. The proposed method for enhancing the contrast of fine image details is based on the idea of combining high-frequency filtering and morphological image processing. The method consists of the following steps: noise suppression in the image, high-frequency filtering, morphological image processing, obtaining the resulting image. As a result of applying the method, the brightness of the contours in the image is enhanced. In the resulting image, all objects will have double outlines. The method was tested in the processing of 50 chest radiographs of patients with various pathologies. Radiographs were performed at the Mariinsky Hospital of St. Petersburg using digital stationary and mobile X-ray machines. In most of the radiographs, it was possible to improve the images contrast, to highlight the objects boundaries. Besides, the method was applied in microfocus X-ray tomography to improve the information content of projection data and improve the reconstruction of the 3D image of the research object. In both the first and second cases, the method showed satisfactory results. The developed method makes it possible to significantly increase the information content of microfocus X-ray images. The obtained practical results make it possible to count on broad prospects for the method application, especially in microfocus X-ray.

Semantic segmentation is a challenging task in computer vision which is widely used in autonomous driving and scene understanding. State‐of‐the‐art semantic segmentation networks, like DeepLab and PSPNet, make full use of multiple feature information to improve spatial resolution. However, the feature resolution in the scale‐axis is not dense enough for practical applications. To tackle this problem, a multi‐stream network is designed with atrous convolutional layers at multiple rates to capture objects and context at multiple scales. Furthermore, intra‐connections and inter‐connections are designed to fuse multi‐scale features densely which produce a feature pyramid with much larger scale diversity and larger receptive field by involving small quantity of computation. The proposed module can be easily used in other methods and it helps to increase the performance. Compared with existing methods, the proposed network, called Multi‐stream Densely Connected Network, reaches competitive results on ADE20K dataset, PASCAL VOC 2012 dataset, and Cityscapes dataset.

Human-inspired neuromorphic computing system breaks the limit of “von Neumann bottleneck” and improves the computing efficiency. However, the processing of a large amount of non-structural visual data results in a considerable waste of computing resources. In human visual system, visual information is preprocessed by the retina before transmitted to the visual cortex, which effectively reduces data redundancy. Therefore, multifunctional intelligent devices which integrate sensing-memory-processing functions are desired for the artificial visual system. Herein, an image memory logic unit (IMLU) inspired by human retina is proposed. The IMLU is formed by a simple two terminal photodetector and comparator module. The photodetector is based on the MoxW1-xS2 alloy with persistent photo conductivity effect. In addition, the IMLU exhibits capabilities of in situ sensing, memory and logical operations (“AND”, “OR”, and “XOR”). Furthermore, the IMLU array enables image processing functions such as dilation, erosion and difference recognition, which shows great potential to provide simplified information processing for subsequent tasks in machine vision.

One of the major issues in today’s agriculture fields is detecting weed plants in between the crops. Weeds consume more water, nutrients, and light compared to crop plants. Being hardy and vigorous in growth habits, they grow way to faster than crops and consume a huge amount of water and nutrients, results causing heavy losses in yields, the process of removal of weeds manually is a difficult job and it requires more manpower. To date, weed removal can’t be automated without manpower. Herbicides play a crucial role in removing the weeds but that leads to soil infertile and later the weeds dominate the field automatically. In solution to reduce the weeds is using herbicide in a higher amount than normal day by day. Usage of herbicides in that amount causes the land infertile. This paper deals with detecting the weeds in the crop using a convolutional neural network, Image processing, and IoT. The weeds in the field and between the crops are detected and removed by using the image processing technique. CNN algorithm is implemented in Matlab software to detect the weed areas in the fields. A robot model is connected to the controller through the motor driver which is also used to carry the camera through the field to detect the weed. The videos and images taken by the camera send to the Matlab and they are trained by using the CNN algorithm and that classifies whether it is a weed or a normal crop. And the necessary instructions send to the Arduino through Zigbee. If the camera detects any weed then the cutter is on 10 seconds to cut the weeds. And the robot model moves further until it finds the next weed. Users can also control the robot model whenever itneeds. Keywords: CNN; Weed cutter; Matlab; Zigbee; Image processing.

Haze and fog are big reasons for road accidents. The haze occurrence in the air lowers the images quality captured by visible camera sensors. Haze brings inconvenience to numerous computer vision applications as it diminishes the scene visibility. Haze removal techniques recuperate the color and scene contrast. These haze removal techniques are extensively utilized in numerous applications like outdoor surveillance, object detection, consumer electronics, etc. Haze removal is commonly performed under the physical degradation model, which requires a solution of an ill-posed inverse issue. Different dehazing algorithms was recently proposed to relieve this difficulty and has acknowledged a great deal of consideration. Dehazing is basically accomplished through four major steps: hazy images acquisition process, estimation process (atmospheric light, transmission map, scattering phenomenon, and visibility or haze level), enhancement process (improved visibility level, reduce haze or noise level), restoration process (restore enhanced image, image reconstruction). This four-step dehazing process makes it possible to provide a step-by-step approach to the complex solution of the ill-posed inverse problem. Our detailed survey and experimental analysis on different dehazing methods that will help readers understand the effectiveness of the individual step of the dehazing process and will facilitate development of advanced dehazing algorithms. The overall objective of this review paper is to explore the various methods for efficiently removing the haze and short comings of the earlier presented techniques used in the revolutionary era of image processing applications.

The rate of road traffic accidents in the 38th and 39th week of 2020 has increased. Many factors influence the occurrence of accidents, one of the reasons is that the driver is sleepy. The parameters used to determine the condition of drowsiness are identifying the condition of the eyelids. With this identification, an image processing system can be utilized by applying the Haar Casecade Classifier and Circular Hough Transform methods with Find Countur using HSV parameters. The detection uses a camera with the driver's eye input, if the image detected is a drowsy eye, the system will wake up the driver by turning on a sound alarm, besides this system is equipped with an SMS gateway feature that will be sent to the driver's relatives, that the driver is sleepy. Message notification sent contains the condition and location of the rider. Based on system-wide testing, this system has a success rate of 100% against HSV1, 96% against HSV2 and HSV3, 89% against HSV4, and 88% against HSV5 on eye detection results and response time. The resulting error is because the lighting at a certain distance is not match with HSV set, the system experiences a freeze due to the buzzer sound delay that is too long and the performance of the PC is also less than optimal.

Face detection is one of the most important subjects in image processing. Over time, researchers have paid much attention to the subject, and they have made tremendous progress in the quality of face detection. In addition to the quality of face detection, the speed of face detection is of prime importance. In this paper, a real-time approach is presented for face detection using the Gabor filters and the neural networks that can be implemented in IoT devices. The Gabor filters are one of the most powerful tools in image processing, but they are rarely used in real-time applications due to high computational complexity. To overcome the problem, a new algorithm is proposed for processing images and detecting faces called circular sliding window (CSW). This new algorithm can reduce the number of sub-images generated by almost 98% related to the sliding window algorithm, in frontal face images which have symmetry. Also, a new Gabor feature called compressed Gabor feature (CGF) is employed which improves the speed of classification due to reducing the size of feature vector of the neural network. In the proposed method, the best speed of face detection and the worst speed of face detection for faces with size of 64 × 64 pixels are 0.0072 and 0.0092 s, respectively. The sensitivity of face detection in the proposed method is 95%, approximately.

Scene analysis in video sequences is a complex task for a computer vision system. Several schemes have been addressed in this analysis, such as deep learning networks or traditional image processing methods. However, these methods require thorough training or manual adjustment of parameters to achieve accurate results. Therefore, it is necessary to develop novel methods to analyze the scenario information in video sequences. For this reason, this paper proposes a method for object segmentation in video sequences inspired by the structural layers of the visual cortex. The method is called Neuro-Inspired Object Segmentation, SegNI. SegNI has a hierarchical architecture that analyzes object features such as edges, color, and motion to generate regions that represent the objects in the scenario. The results obtained with the Video Segmentation Benchmark VSB100 dataset demonstrate that SegNI can adapt automatically to videos with scenarios that have different nature, composition, and different types of objects. Also, SegNI adapts its processing to new scenario conditions without training, which is a significant advantage over deep learning networks.

As we all know, detection and recognition of distant or relatively small targets is a challenging problem. However, the eagle-eye is known for its sharp eyesight, which is called clairvoyance. Inspired by the structural characteristics and physiological mechanism of eagle-eye, a lightweight vision system is designed for target detection and recognition in this paper. First, a hardware platform is established, which consists of a pan-tilt, a short-focus camera and a long-focus camera. Then, combined with the image processing mechanism of the eagle-eye and deep learning, the cameras with different focal lengths are controlled cooperatively to achieve target searching, edge extraction, target detection and recognition. Finally, by separately detecting the targets with different focal length cameras, it can be found that the confidence of the target detection with the long-focus camera is greatly improved. Experimental results show that the designed lightweight eagle-eye vision system can accurately detect and recognize targets, which has a strong adaptive ability.

In order to reduce the harmfulness of explosion crime to society, this paper improves the coordinate transformation algorithm from the perspective of bomb demolition, and builds a motion control system suitable for EOD robots. Moreover, this article combines image recognition algorithms to improve machine vision, and obtains an intelligent recognition system suitable for EOD robots. In addition, the optical model of the monocular camera is established to calibrate the camera's own parameters, and a series of image processing algorithms are used for visual image processing to obtain the plane position information of the target object. After the robot recognizes the position of the bomb, it gradually approaches the bomb, its coordinates are constantly changing, and the collected data is transmitted to the data processing center to realize real-time control of the EOD robot. Finally, this paper verifies the reliability of the EOD robot through simulation tests.

Infrared and visible image fusion is an effective image processing technique to obtain more comprehensive information, which can help people better understand various scenarios. In this paper, a novel infrared and visible image fusion method is proposed which fully considers the attributes of objects in source images. Benefiting from the attribute and the edge-preserving filters, the prominent objects in the infrared source image are effectively extracted. Then, the weight-based Laplacian pyramid fusion strategy is adopted to get more natural fusion results. The experimental results on the public image fusion datasets and a new infrared–visible video fusion dataset show that the proposed method achieves state-of-the-art fusion performance in terms of both visual and objective evaluations. The proposed algorithm is also implemented in an infrared–visible dual sensor system, which demonstrated the practicability of our fusion method.

Background: With the rapid development of automobile technology, the problem of abnormal door-closing noise has attracted more and more attention. The abnormal door-closing noise is an important factor for judging the quality of a car, so accurate identification of abnormal noise is the premise of fundamentally solving the abnormal noise. Objective: To accurately identify the abnormal sound of car closing through the image processing method. Methods: To accurately identify the abnormal noise of car closing the method to recognize and classify images using Support Vector Machine (SVM) is proposed. This method extracts Histogram of Oriented Gradient (HOG), texture and Speed Up Robust Features (SURF). The three extracted feature vectors are combined and used as the input of SVM. The classifiers obtained by different kernel functions are used to predict the labels of the test set, respectively. Results: Calculating the ratio of the value on the diagonal of the confusion matrix to the total number of each row, and this ratio is the classification accuracy rate. Test accuracy rate is 85%, the results indicate that the accuracy is high. Conclusion: This paper uses image processing methods to extract HOG, GLCM, SURF features and merge them together as a new feature vector. The experimental results show that the SVM classifier using the Gaussian kernel function optimized by hyperparameters has a high accuracy rate and can be used to identify whether the door is closed with abnormal noise.

Motion blur is a common artifact in image processing, specifically in e-health services, which is caused by the motion of a camera or scene. In linear motion cases, the blur kernel, i.e., the function that simulates the linear motion blur process, depends on the length and direction of blur, called linear motion blur parameters. The estimation of blur parameters is a vital and sensitive stage in the process of reconstructing a sharp version of a motion blurred image, i.e., image deblurring. The estimation of blur parameters can also be used in e-health services. Since medical images may be blurry, this method can be used to estimate the blur parameters and then take an action to enhance the image. In this paper, some methods are proposed for estimating the linear motion blur parameters based on the extraction of features from the given single blurred image. The motion blur direction is estimated using the Radon transform of the spectrum of the blurred image. To estimate the motion blur length, the relation between a blur metric, called NIDCT (Noise-Immune Discrete Cosine Transform-based), and the motion blur length is applied. Experiments performed in this study showed that the NIDCT blur metric and the blur length have a monotonic relation. Indeed, an increase in blur length leads to increase in the blurriness value estimated via the NIDCT blur metric. This relation is applied to estimate the motion blur. The efficiency of the proposed method is demonstrated by performing some quantitative and qualitative experiments.

With today's advancing technology, support of developing hardware and software systems, the developments in the field of medicine have increased considerably. In particular, medical image analysis and processing systems have taken a considerable lead. The development of an automatic system could provide great convenience for doctors and practitioners in the field. The image processing techniques proposed in this study can contribute to more effective analysis and more accurate diagnosis, regardless of the individual levels of experience of the users or particular situations and conditions such as fatigue or image quality. This paper presents a robust method for retinal blood vessel segmentation and some automatic algorithms for analyzing the vessel network and pixel classification into vessel and non-vessel .The aim of the work extraction or segmentation of retinal blood vessels used computer vision and image processing for getting high accuracy (comparison with manual) .Also We used the preprocessing techniques for enhancement of the image .And used two types of filter for comparing result to get best scenario. Also for simulation result we used the matlab and implement on DRIVE database.

Human visual perception is limited to the visual-optical spectrum. Machine vision is not. Cameras sensitive to the different infrared spectra can enhance the abilities of autonomous system...

Purpose The purpose of this paper is to monitor the backup indicators in case of indicator failure and to minimize the situations when the pilot may be unable to monitor the indicator effectively in emergency situations. Design/methodology/approach In this study, the pointer positions of different indicators were determined with a deep learning-based algorithm. Within the scope of the study, the pointer on the analog indicators obtained from aircraft cockpits was detected with the YOLOv4 object detector. Then, segmentation was made with the GrabCut algorithm to detect the pointer in the detected region more precisely. Finally, a line including the segmented pointer was found using the least-squares method, and the exact direction of the pointer was determined and the angle value of the pointer was obtained by using the inverse tangent function. In addition, to detect the pointer of the YOLOv4 object detection method and to test the designed method, a data set consisting of videos taken from aircraft cockpits was created and labeled. Findings The analog indicator pointers were detected with great accuracy by the YOLOv4 and YOLOv4-Tiny detectors. The experimental results show that the proposed method estimated the angle of the pointer with a high degree of accuracy. The developed method can reduce the workloads of both pilots and flight engineers. Similarly, the performance of pilots can be evaluated with this method. Originality/value The authors propose a novel real-time method which consists of detection, segmentation and line regression modules for mapping the angle of the pointers on analog indicators. A data set that includes analog indicators taken from aircraft cockpits was collected and labeled to train and test the proposed method.

Camshift algorithm tracking is susceptible to interference when a tracking object is occluded or when its hue is similar to the background. An improved Camshift object-tracking algorithm combining AKAZE (Accelerated-KAZE) feature matching and Kalman filtering is proposed. First, the video channel is converted for processing. Second, AKAZE is used to match the object feature points and Kalman filtering is used to predict the next position. Then different scenes are judged by the threshold and the Camshift and Kalman tracking algorithms are used for object tracking, respectively. Finally, the improved Camshift algorithm is used to test the moving object in a variety of situations and compared with the traditional Camshift algorithm and the Kalman filter improved Camshift algorithm. Experimental results show that the improved joint tracking algorithm can continue tracking under full occlusion. The effective frame rate of recognition is increased by about 20%, and the single-frame image processing time is less than 35 ms, which can meet the real-time tracking requirements.

The visibility and analyzability of MRI and CT images have a great impact on the diagnosis of medical diseases. Therefore, for low-quality MRI and CT images, it is necessary to effectively improve the contrast while suppressing the noise. In this paper, we propose an enhancement and denoising strategy for low-quality medical images based on the sequence decomposition Retinex model and the inverse haze removal approach. To be specific, we first estimate the smoothed illumination and de-noised reflectance in a successive sequence. Then, we apply a color inversion from 0-255 to the estimated illumination, and introduce a haze removal approach based on the dark channel prior to adjust the inverted illumination. Finally, the enhanced image is generated by combining the adjusted illumination and the de-noised reflectance. As a result, improved visibility is obtained from the processed images and inefficient or excessive enhancement is avoided. To verify the reliability of the proposed method, we perform qualitative and quantitative evaluation on five MRI datasets and one CT dataset. Experimental results demonstrate that the proposed method strikes a splendid balance between enhancement and denoising, providing performance superior to that of several state-of-the-art methods.

Traditional bridge inspection methods mostly rely on the experience and expertise of inspectors in visual detection of cracks. Such visual inspection methods are not only time consuming but also could be affected by subjective judgment of inspectors and lack a unified inspection standard. To overcome the shortcomings of visual inspection, this study investigates the feasibilities of using an unmanned aerial vehicle (UAV) combined with laser projectors to measure surface cracks. All related parameters of the laser projectors that are fixed on the UAV are calibrated first in the laboratory. Then, the UAV with the laser projectors is used to capture images of surface cracks. The images from the UAV can be transformed to orthoimages using the laser projectors’ parameters and the image processing technology. Next, surface cracks in the orthoimages are automatically identified by a crack identification algorithm. Finally, the characteristics of surface cracks are evaluated by the image-based measurement technologies. To verify the accuracy of the proposed image-based measurement system (UAV with laser projectors), seven artificial cracks with different widths are measured using the proposed measurement system and the point cloud method (UAV with the global navigation satellite system) under different measurement distances. The test results demonstrate that while the distance between the UAV and the artificial crack is within 150 cm, the proposed image-based measurement system is able to measure crack widths more accurately than the GNSS point cloud method. In addition, the proposed image-based measurement system can automatically identify surface cracks and measure crack widths from UAV-captured images without manual measurements.

Cracks in road surfaces are among the most critical problems in the management of expressways. Their sizes, which not only have severe consequences in relation to expressway safety, decide maintenance timing, which if accurately estimated, can prevent the rapid development of crack disease. Currently, the main techniques used for crack size measurement makes use of either human inspection or sophisticated mobile equipment, which are respectively, very time consuming, labour-intensive, and difficult to implement in civil engineering. Thus, there is great interest in practical and highly efficacious automated crack identification systems coupled with low-cost data-processing sensors. A PTZ (pan/tilt/zoom) camera-based image processing idea is investigated in this paper for crack size measurement in expressways in China. The proposed method consists of four steps: 1) image acquisition, 2) corner-based line size calculation, 3) pixel size calculation, and 4) crack size estimation. A case study of roadway lane images and crack images, collected from China G4/Jingshi, is conducted to demonstrate the feasibility of the proposed method. The results demonstrate that PTZ camera-based image processing is capable of efficiently quantifying cracks, obtaining high accuracy crack size results at range of 0 m to 40 m. The method shows promise for crack size measurement in a cost-effective manner and can provide an objective and timely means for transportation agencies to conduct maintenance plans.