Search

An image through the digitization process is referred to as a digital image. The quality of the digital image may be degenerating due to interferences on the acquisition, transmission, extraction, etc. This attracted the attention of many researchers to study the causes of damage to the information in the image. In addition to finding cause of image damage, the researchers also looking for ways to overcome this problem. There are many filtering techniques that have been introduced to deal the damage to the information in the image. In addition to eliminating noise from the image, filtering techniques also aims to maintain the originality of the features in the image. Among the many research papers on image filtering there is a lack of review papers which are an important to facilitate researchers in understanding the differences in each filtering technique. Additionally, it helps researchers determine the direction of research conducted based on the results of previous research. Therefore, this paper presents a review of several filtering techniques that have been developed so far. [ABSTRACT FROM AUTHOR]

One of the most important difficulties which doctors face in diagnosing is the analysis and diagnosis of brain stroke in magnetic resonance imaging (MRI) images. Brain stroke is the interruption of blood flow to parts of the brain that causes cell death. To make the diagnosis easier for doctors, many researchers have treated MRI images with some filters by using Matlab program to improve the images and make them more obvious to facilitate diagnosis by doctors. This paper introduces a digital system using hardware concepts to clarify the brain stroke in MRI image. Field programmable gate arrays (FPGA) is used to implement the system which is divided into four phases: preprocessing, adjust image, median filter, and morphological filters alternately. The entire system has been implemented based on Zynq FPGA evaluation board. The design has been tested on two MRI images and the results are compared with the Matlab to determine the efficiency of the proposed system. The proposed hardware system has achieved an overall good accuracy compared to Matlab where it ranged between 90.00% and 99.48%. [ABSTRACT FROM AUTHOR]

Deep learning has gained high popularity in the field of image processing and computer vision applications due to its unique feature extraction property. For this characteristic, deep learning networks used to solve different issues in computer vision applications. In this paper the issue has been raised is classification of logo of formal directors in Iraqi government. The paper proposes a multi-layer convolutional neural network (CNN) to classify and recognize these official logos by train the CNN model on several logos. The experimental show the effectiveness of the proposed method to recognize the logo with high accuracy rate about 99.16%. The proposed multi-layers CNN model proves the effectiveness to classify different logos with various conditions. [ABSTRACT FROM AUTHOR]

Numerous technologies and systems, including autonomous vehicles, surveillance systems, and robotic applications, rely on the capability to accurately detect pedestrians to ensure their safety. As the demand for realtime object detection continues to rise, many researchers have dedicated their efforts to developing effective and trustworthy algorithms for pedestrian recognition. By integrating learning complexity-aware cascades with an enhanced you only look once (YOLO) algorithm, the paper presents a real-time system for identifying both items and pedestrians. The performance of the proposed approach is evaluated using the Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI) pedestrian dataset across both the v4 and v8 versions of the YOLO framework. Prioritizing both speed and accuracy, the enhanced YOLO algorithm outperforms its baseline counterpart. The demonstrated superiority of the suggested technique on the KITTI pedestrian dataset underscores its effectiveness in real-world contexts. Furthermore, the complexity-aware learning cascades contribute to a streamlined detection model without compromising performance. When applied to scenarios requiring real-time identification of objects and individuals, the proposed method consistently delivers promising outcomes. [ABSTRACT FROM AUTHOR]

For the single spectrum image could not fully express the target feature information, this paper proposed a multispectral image fusion method based on non-subsample contourlet transform (NSCT). For the low frequency coefficients decomposed, fourth-order correlation coefficient is used to calculate the correlation between each low frequency coefficients, averaging fusion for the higher correlation coefficient, weight phase congruency fusion for the low correlation coefficient. For high frequency coefficients, Gaussian weight sum modified Laplace method is used for fusing, to retain more local structure details. Simulation results show that the method effectively retain the image structure information and more local details, and increase the image contrast. [ABSTRACT FROM AUTHOR]

Due to the rapid development of information technology, the process of exchanging information over the web has become simple and fast. Nevertheless, the transmission of information will be at risk due to the nature of the internet, as anyone can access and change the data by hacking it. Therefore, it is necessary for protecting any personal information from being accessed by unlawful individuals. It became necessary to use encryption and steganography techniques for protecting data. Steganography is a manner used for covering hidden information in several other media while preserving the form of the original data and leaving no evidence of hidden data, whereas, encryption changes the original information into unclear or ambiguous information, which is called scrambled. The paper presents a method of video steganography as a powerful and effective tool for data steganography. We used a video as a spreading medium and then used it for concealment of an audio message and a document (in a pdf file) in a method that makes the information invisible using the least significant bit (LSB) method. [ABSTRACT FROM AUTHOR]

Image fusion is a comprehensive information processing technique and its purpose is to enhance the reliability of the image via the processing of the redundant data among multiple images, improve the image definition and information content through fusion of the complementary information of multiple images so as to obtain the information of the objective or the scene in a more accurate, reliable and comprehensive manner. This paper uses the sparse representation method of compressive sensing theory, proposes a multi-source and multi-feature image information fusion method based on compressive sensing in accordance with the features of image fusion, performs sparsification processing on the source image with K-SVD algorithm and OMP algorithm to transfer from spatial domain to frequency domain and decomposes into low-frequency part and high-frequency park. Then it fuses with different fusion rules and the experimental results prove that the method of this paper is better than the traditional methods and it can obtain better fusion effects. [ABSTRACT FROM AUTHOR]

The image is usually polluted by noises in its acquisition and transmission and noises are of great importance in the image quality, therefore, image de-noising has become a significant technique in image analysis and processing. In the image de-noising based on sparse representation, one of the hot spots in recent years, the useful image information has certain structural features, which coincide with the atomic structure while noises don't have such features, therefore, sparse representation can separate the useful information from the noises effectively so as to achieve the purpose of de-noising. In view of the abovementioned theoretical basis, this paper proposes an image de-noising algorithm of sparse representation based on K-means Singular Value Decomposition (K-SVD). This method can integrate the construction and optimization of over-complete dictionary, train the atom dictionary with the image samples to be decomposed and effectively build the atom dictionary that reflects various image features to enhance the de-noising performance of the algorithm in this paper. Through simulation analysis, this method can conduct noise filtration on the image with different noise densities and its de-noising effect is also better than other methods. [ABSTRACT FROM AUTHOR]

Developing hotspot prediction models using decision tree algorithms require target classes to which objects in a dataset are classified. In modeling hotspots occurrence, target classes are the true class representing hotspots occurrence and the false class indicating non hotspots occurrence. This paper presents the results of satellite image processing in order to determine the radius of a hotspot such that random points are generated outside a hotspot buffer as false alarm data. Clustering and majority filtering were performed on the Landsat TM image to extract burn scars in the study area i.e. Rokan Hilir, Riau Province Indonesia. Calculation on burn areas and FIRMS MODIS fire/hotspots in 2006 results the radius of a hotspot 0.90737 km. Therefore, non-hotspots were randomly generated in areas that are located 0.90737 km away from a hotspot. Three decision tree algorithms i.e. ID3, C4.5 and extended spatial ID3 have been applied on a dataset containing 235 objects that have the true class and 326 objects that have the false class. The results are decision trees for modeling hotspots occurrence which have the accuracy of 49.02% for the ID3 decision tree, 65.24% for the C4.5 decision tree, and 71.66% for the extended spatial ID3 decision tree. [ABSTRACT FROM AUTHOR]

In the image acquisition or transmission, the image may be damaged and distorted due to various reasons; therefore, in order to satisfy people's visual effects, these images with degrading quality must be processed to meet practical needs. Integrating artificial bee colony algorithm and fuzzy set, this paper introduces fuzzy entropy into the self-adaptive fuzzy enhancement of image so as to realize the selfadaptive parameter selection. In the meanwhile, based on the exponential properties of information increase, it proposes a new definition of fuzzy entropy and uses artificial bee colony algorithm to realize the self-adaptive contrast enhancement under the maximum entropy criterion. The experimental result shows that the method proposed in this paper can increase the dynamic range compression of the image, enhance the visual effects of the image, enhance the image details, have some color fidelity capacity and effectively overcome the deficiencies of traditional image enhancement methods. [ABSTRACT FROM AUTHOR]

Copyright of Telkomnika is the property of Department of Electrical Engineering, Ahmad Dahlan University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this paper, we propose a novel and robust local image descriptor to resolve the problem of complex illumination variations on low-level feature matching. Many classic local feature descriptors are invariant to linear illumination or monotonous intensity shift, but cannot handle more complex nonlinear illumination changes, which often occur due to the different time of exposure, the viewing direction, different types of light-surface interactions and other varying brightness changes. The presented descriptor that extracts the histograms of oriented gradients and Uniform Symmetric-Local Binary Pattern (US-LBP) features for each feature point neighborhood in the scale space. Extensive experiments show that the proposed descriptor outperforms many state-of-the-art descriptors such as SIFT, ORB, SURF and FREAK under the problem of local image matching, especially demonstrating the effectiveness of our method under complex illumination changes. [ABSTRACT FROM AUTHOR]

For the defect in describing affine and blur invariable of scale-invariant feature transform (SIFT) at large viewpoint variation, a new object recognition method is proposed in this paper, which used maximally stable extremal region (MSER) detecting MSERs and SIFT describing local feature of these regions. First, a new most stability criterion is adopt to improve the detection effect at irregular shaped regions and under blur conditions; then, the local feature descriptors of MSERs is extracted by the SIFT; and finally, the method proposed is comparing then correct rate of SIFT and the proposed through image recognition with standard test images. Experimental results show that the method proposed can still achieve more than 74% recognition correct rate at different viewpoint, which is better than SIFT. [ABSTRACT FROM AUTHOR]

In the last years, computer vision systems for the detection of skin cancer have been proposed, especially using machine learning techniques for the classification of the disease and features based on the ABCD dermatology criterion, which gives information on the status of the skin lesion based on static properties such as geometry, color, and texture, making it an appropriate criterion for medical diagnosis systems that work through images. This paper proposes a novel skin cancer classification system that works on images taken from a standard camera and studies the impact on the results of the smoothed bootstrapping, which was used to augment the original dataset. Eight classifiers with different topologies (KNN, ANN, and SVM) were compared, with and without data augmentation, showing that the classifier with the highest performance as well as the most balanced one was the ANN with data augmentation, achieving an AUC of 87.1%, which saw an improvement from an AUC of 84.3% of the ANN trained with the original dataset. [ABSTRACT FROM AUTHOR]

Physiological maturity is an important indicator for beef quality. In traditional method, the maturity grade is determined by subjectively evaluating the degree of cartilage ossification at the tips of the dorsal spine of the thoracic vertebrae. This paper uses the computer vision to replace the artificial method for extracting object (cartilage and bone) regions. Hu invariant moments of object region were calculated as the regional shape characteristic parameters. A trained Hopfield neural network model was used for recognizing cartilage and bone area in thoracic vertebrae image based on minimum Euclidean distance. The result showed that the accuracy of network recognition for cartilage and bone region was 92.75% and 87.68%, respectively. For automatically maturity prediction, the accuracy of prediction was 86%. Algorithm proposed in this paper proved the image description and neural network modeling was an effective method for extracting image feature regions. [ABSTRACT FROM AUTHOR]

This paper proposes image compression using discrete cosine transform (DCT) for the format of joint photographic expert groups (JPEG) or JPG, portable network graphic (PNG) and bitmap (BMP). These three extensions are the most popular types used in current image processing storage. The purpose of image compression is to produce lower memory usage or to reduce memory file. This process removes redundant information of each pixel. The challenge for image compression process is to maintain the quality of images after the compression process. Hence, this article utilizes the DCT technique to sustain the image quality and at the same time reduces the image storage size. The effectiveness of the DCT technique has been reasonable over some real images and the implementation of the technique has been compared with different types of image extensions. Matlab software is an important platform for this project in order to write a program and perform the progress of project phase by phase to achieve the expected results. Based on the tested images, the DCT technique in image compression is capable to reduce the image storage memory in average about 50% of each image tested. [ABSTRACT FROM AUTHOR]

This paper proposes, a novel hybrid color component (HCC) issued from amounts number of color space with iterative manner, in fact traditional images obtained by RGB sensor weren't the effective way in image processing applications, for this purpose we have propose a supervised algorithm to substitute RGB level by hybrid and suitable color space at the aim to make well representation of the handled amounts of data, this step is extremely important because the obtained results it will be injected in many future studies like tracking, classification, steganography and cryptography. The second part of this paper consists to segment image coded in hybrid color space already selected, the used algorithm is inspired from kernel function where statistical distribution was used to model background and Bayes rule to make decision of the membership of each pixel, in this research topics we have extended this algorithm in the aim to improve compactness of these distribution. Cauchy background modeling and subtraction is used, and shows the high accuracy of automatic player detection. [ABSTRACT FROM AUTHOR]

Personal identification has become one of the most important terms in our society regarding access control, crime and forensic identification, banking and also computer system. The fingerprint is the most used biometric feature caused by its unique, universality and stability. The fingerprint is widely used as a security feature for forensic recognition, building access, automatic teller machine (ATM) authentication or payment. Fingerprint recognition could be grouped in two various forms, verification and identification. Verification compares one on one fingerprint data. Identification is matching input fingerprint with data that saved in the database. In this paper, we measure the performance of the memetic algorithm to process the image fingerprints dataset. Before we run this algorithm, we divide our fingerprints into four groups according to its characteristics and make 15 specimens of data, do four partial tests and at the last of work we measure all computation time. [ABSTRACT FROM AUTHOR]

Copyright of Telkomnika is the property of Department of Electrical Engineering, Ahmad Dahlan University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The limitation of traditional iris recognition systems to process iris images captured in unconstraint environments is a breakthrough. Automatic iris recognition has to face unpredictable variations of iris images in real-world applications. For example, the most challenging problems are related to the severe noise effects that are inherent to these unconstrained iris recognition systems, varying illumination, obstruction of the upper or lower eyelids, the eyelash overlap with the iris region, specular highlights on pupils which come from a spot of light during captured the image, and decentralization of iris image which caused by the person's gaze. Iris segmentation is one of the most important processes in iris recognition. Due to the different types of noise in the eye image, the segmentation result may be erroneous. To solve this problem, this paper develops an efficient iris segmentation algorithm using image processing techniques. Firstly, the outer boundary segmentation of the iris problem is solved. Then the pupil boundary is detected. Testes are done on the Chinese Academy of Sciences' Institute of Automation (CASIA) database. Experimental results indicate that the proposed algorithm is efficient and effective in terms of iris segmentation and reduction of time processing. The accuracy results for both datasets (CASIA-V1 and V4) are 100% and 99.16 respectively. [ABSTRACT FROM AUTHOR]

The process of identifying images and patterns is one of the most important processes of digital image processing, which is used in many applications such as fingerprint recognition, face recognition and pattern recognition. Due to the large size of the image, the process of identifying the image requires a great time, which in turn leads us to extract some characteristics of the magnitude of the volume, which can be used as an identifier to retrieve the image or recognize it and thus we have devoted a lot of time to identify the image. In this research paper, a modified symmetric local binary pattern (MSLBP) method was proposed to extract texture features. The proposed algorithm was implemented on many digital fingerprint's images and the local structure features of these images were obtained. Several image recognition experiments are conducted on these features and compared with other algorithms. The results of the proposed algorithm showed that the digital image was represented in a very small size and furthermore the speed and accuracy of image recognition based on the proposed method was increased significantly. Unlike the methods based on LBP, the proposed method gives the same features of the image even if the image was rotated with any angle [ABSTRACT FROM AUTHOR]

Automated medical image processing, particularly of radiological images, can reduce the number of diagnostic errors, increase patient care and reduce medical costs. This paper seeks to evaluate the performance of three recent convolutional neural networks in the autonomous identification of fissures over two-dimensional radiological images. These architectures have been proposed as deep neural network types specially designed for image classification, which allows their integration with traditional image processing strategies for automatic analysis of medical images. In particular, we use three convolutional networks: ResNet (residual neural network), DenseNet (dense convolutional network), and NASNet (neural architecture search network) to learn information from a set of 200 images labeled half as fissured bones and half as seamless bones. All three networks are trained and adjusted under the same conditions, and their performance was evaluated with the same metrics. The final results consider not only the model's ability to predict the characteristics of an unknown image but also its internal complexity. The three neural models were optimized to reduce classification errors without producing network over-adjustment. In all three cases, generalization of behavior was observed, and the ability of the models to identify the images with fissures, however the expected performance was only achieved with the NASNet model. [ABSTRACT FROM AUTHOR]

Digital image is considered as a powerful tool to carry and transmit information between people. Thus, it attracts the attention of large number of researchers, among them those interested in preserving the image features from any factors that may reduce the image quality. One of these factors is the noise which affects the visual aspect of the image and makes others image processing more difficult. Thus far, solving this noise problem remains a challenge for the researchers in this field. A lot of image denoising techniques have been introduced in order to remove the noise by taking care of the image features; in other words, getting the best similarity to the original image from the noisy one. However, the findings are still inconclusive. Beside the enormous amount of researches and studies which adopt several mathematical concepts (statistics, probabilities, modeling, PDEs, wavelet, fuzzy logic, etc.), there is also the scarcity of review papers which carry an important role in the development and progress of research. Thus, this review paper intorduce an overview of the different fundamental approaches that yield the several image-denoising techniques, presented with a new classification. Furthermore, the paper presents the different evaluation tools needed on the comparison between these techniques in order to facilitate the processing of this noise problem, among a great diversity of techniques and concepts. [ABSTRACT FROM AUTHOR]

Since the concept of coverless information hiding was proposed, it has been greatly developed due to its effectiveness of resisting the steganographic tools. In this paper, a new coverless steganography is presented to hide the secret data in a more secure way and to enhance the robustness against attacks. This method depends on frequency domain. The embedding process consists of several steps. Firstly, the secret data is divided into no overlapping segments. Secondly, a set of images is collected to find appropriate images to be stego images. Thirdly, to build a hash sequence for an image, a powerful hashing algorithm is used. Fourthly, for each image hash sequence, the inverted index structure is created. Fifthly, choose the image which its hash equivalent to the secret data segment. Several tests are done to measure the robustness of the proposed method. The results of the experiments reveal that the proposed strategy is resistant to a variety of image processing attacks such as joint photographic experts group (JPEG) compression, noise, low pass filtering, scaling, rotation and median and mean filter, brightness, and sharpening. [ABSTRACT FROM AUTHOR]

In the process of production of shampoo, the smoothness of the bottle cap is an important part of the quality inspection. At present, the Cap smoothness detect is mainly completed by the human eye, the method has large error and accidental, is not precise enough. In this paper, a method based on image processing to detect the smoothness of bottle cap is proposed. The hardware device is designed and set up to realize the matching of image acquisition and light source. The sliding platform based on parallel slide rail is designed, which can be used to enter and remove samples, and to realize the accurate positioning of the sample. MATLAB and C language programming are used to achieve the image segmentation, enhancement, filtering and other processing of the bottle cap. This system overcomes the problem of eye detection, effectively eliminates the influence of outside interference, and achieves a high accuracy. [ABSTRACT FROM AUTHOR]

Polarimetric SAR (PolSAR) has played more and more important roles in earth observation. Polarimetric SAR image classification is one of the key problems in the PolSAR image interpretation. In this paper, based on the scattering properties of fully polarimetric SAR data, combing the statistical characteristics and neighborhood information, an efficient unsupervised method of fully polarimetric SAR image classification is proposed. In the method, polarimetric scattering characteristics of fully polarimetric SAR image is used, and in the denoised total power image of polarimetric SAR, SPAN (the total polarimetric power), the texture features of gray level co–occurrence matrix are extracted at the same time. Finally, the polarimetric information and texture information are combined for fully polarimetric SAR Image classification with clustering algorithm. The experimental results show that better classification results can be obtained in the Radarsat-2 data with the proposed method. [ABSTRACT FROM AUTHOR]

In order to accomplish the automatic nondestructive testing, a parts surface structure image classification detection system is designed. A series of parts surface texture images have been obtained from different processing methods for feature analysis and the combination of pre-processing method by MATLAB image processing toolbox has been put forward, using statistical analysis method for feature extraction. Based on the established BP neural network training optimization identification system, this paper realized the recognition of parts surface resulted from four kinds of processing methods: turning, milling, planning and grinding. The research results show that the deficit value of gray level co-occurrence matrix and the histogram matrix variance value can be regarded as characteristic parts of the surface texture structure value, providing foundations for further development of parts surface structure detection. [ABSTRACT FROM AUTHOR]

A novel medical image enhancement algorithm based on spatial domain is presented in this paper. The medical image is firstly divided into several sub-images under the dyadic wavelet scale analysis. As different directional sub-images can reflect different characteristics, at each level, different Butterworth homomorphic filtering functions are used in performing filtering of corresponding sub-band images to attenuate the low frequencies as well as amplify the high frequencies and a linear adjustment is carried out on the low frequency of the highest level. Later, the wavelet reconstruction course is utilized to get the final image. Experiments on magnetic resonance (MR) images of temporomandibular joint (TMJ) soft tissues have shown that our method can eliminate non-uniformity luminance distribution of medical images effectively and its performance is better than traditional Butterworth homomorphic filtering algorithm. [ABSTRACT FROM AUTHOR]

Primary challenges are the identification, segmentation, and extraction of the afflicted area from the scanning of magnetic resonance. However, it is a time-consuming and tiresome for clinical specialists. In this paper, an automated brain tumor system is proposed. The proposed system employs hybrid image processing techniques such as contrast correction, histogram normalization, thresholding techniques, arithmetic, and morphological operations to quarantine nearby organs and other tissue from the brain for improving the localization of the affected region. At first, the skull stripping process is proposed to segregate the non-designated regions to extract the designated brain regions. Those resultant brain region images are further subjected to discover the brain tumor. The planned scheme is studied on the magnetic resonance (MR) images with the use of T1, T2, T1c, and fluid-attenuated inversion recovery (FLAIR). The proposed hybrid method employed. The results reveal that the proposed method is quite efficient to extract the tumor region. The accuracy rate for segmentation and separation of area of interest in brain tumor reached to 95%. Finally, the significance of the proposed procedure is confirmed using the real image clinical dataset got from ten patients were diagnosed as begin, malignant, and metastatic brain tumors in Al-Yarmouk and Baghdad teaching hospital in Baghdad, Iraq. [ABSTRACT FROM AUTHOR]

The color image segmentation is one of most crucial application in image processing. It can apply to medical image segmentation for a brain tumor and skin cancer detection or color object detection on CCTV traffic video image segmentation and also for face recognition, fingerprint recognition etc. The color image segmentation has faced the problem of multidimensionality. The color image is considered in five-dimensional problems, three dimensions in color (RGB) and two dimensions in geometry (luminosity layer and chromaticity layer). In this paper the, L*a*b color space conversion has been used to reduce the one dimensional and geometrically it converts in the array hence the further one dimension has been reduced. The a*b space is clustered using genetic algorithm process, which minimizes the overall distance of the cluster, which is randomly placed at the start of the segmentation process. The segmentation results of this method give clear segments based on the different color and it can be applied to any application. [ABSTRACT FROM AUTHOR]

Image registration, as one of the basic tasks of image processing, is the process to register two images about the same objective or background which are acquired in different times, different sensors, different perspectives and different shooting conditions. In the image registration, because of the problems that the image information is complicated, they have strong correlation and incompleteness, inaccuracy and non-construction occur in different levels in the processing, to apply the method of computational intelligence information processing in the image registration can have better results than the traditional computation methods. This paper proposes an image registration method based on wavelet decomposition and ant colony optimization, which divides the process of image registration into coarse registration and refined registration through wavelet decomposition technique. In the coarse registration, the transformation parameter value of the image approximation component is acquired through ant colony optimization while the changing parameter value of the original image is obtained by the ant colony search method in the refined registration. The simulation experiment shows that this registration method has the characteristics of anti-noise, fast speed, high accuracy and high registration success rate. [ABSTRACT FROM AUTHOR]

Copyright of Telkomnika is the property of Department of Electrical Engineering, Ahmad Dahlan University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Recently, advances in sensor technologies, data communication paradigms, and data processing algorithms all affect the feasibilities of the bridges structural health monitoring and deterioration detection, and other implementations of monitoring operations. The paper proposes a method to design an irregularity detection and monitoring system for road bridges that combines internet of things (IoT) and artificial intelligence (AI) technologies. Raspberry Pi 4 embedded computer integrating IoT and AI technology with convolutional neural network (CNN) is employed to simultaneously monitor remote bridges on websites and apps via Google Firebase cloud database. The first step of successful testing in the laboratory showed that the system can work stably and coincide with the proposed goals. [ABSTRACT FROM AUTHOR]

Copyright of Telkomnika is the property of Department of Electrical Engineering, Ahmad Dahlan University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Telkomnika is the property of Department of Electrical Engineering, Ahmad Dahlan University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Telkomnika is the property of Department of Electrical Engineering, Ahmad Dahlan University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Aedes aegypti mosquitoes are a small slender fly insect that spreads the arbovirus from flavivirus vector through its sucking blood. An early detection of this species is very important because once these species turn into adult mosquitoes a population control becomes more complicated. Things become worse when difficult access places like water storage tank becomes one of the breeding favorite places for Aedes aegypti mosquitoes. Therefore, there is a need to help the field operator during the routine inspection for an automated identification and detection of Aedes aegypti larvae, especially at difficult access places. This paper reviews different methodologies that have been used by various researchers in identifying and counting Aedes aegypti. The objective of the review was to analyze the techniques and methods in identifying and counting the Aedes Aegypti larvae of various fields of study from 2008 and above by taking account their performance and accuracy. From the review, thresholding method was the most widely used with high accuracy in image segmentation followed by hidden Markov model, histogram correction and morphology operation region growing. [ABSTRACT FROM AUTHOR]

Sparse representation-based techniques have shown great potential for pattern recognition problems. Therefore, on the basis of the sparse characteristics of the features for PolSAR image classification, a supervised PolSAR image classification method based on sparse representation is proposed in this paper. It works by projecting the feature vector of the pending pixel onto a subset of training vectors from dictionary and then obtains the corresponding optimal coefficients as well as the residual error with respect to each atom. Then, the residual errors of the pending pixel with respect to each atom are evaluated and considered as the criteria for classification, namely, the ultimate class can be obtained according to the atoms with the least residual error. The verified experiment is implemented using Danish EMISAR L-band fully PolSAR data of Foulum Area (DK) to validate the performance of the proposed classification method.The preliminary experimental results confirm that the proposed method outputs an excellent result and moreover the classification process is simpler and less time consuming. [ABSTRACT FROM AUTHOR]

Video surveillance systems with image recognition and image tracking capabilities have a wide range of applications in the field of city protection, social security and public safety. In this paper, based on the analysis of image detection and tracking principles proposed image recognition based PID fuzzy control and tracking algorithm. First, the image on the basis of the gray-scale image acquired by the video processor, the binary processing, and then through the expansion operation of mathematical morphology the candidate region of the image recognition, and then the final image region obtained through the vertical positioning and horizontal positioning of the region Finally, through the PID control algorithm it is the image of the track. Simulation results show that the PID fuzzy control-based image recognition and tracking algorithm in the image of the vehicle license plate image recognition and tracking performance and good practicability. [ABSTRACT FROM AUTHOR]

This paper presents a multiple objects recognition method using the SURF and the SIFT algorithm. Both algorithms are used for finding features by detecting keypoints and extracting descriptors on every object. The randomized KD-Tree algorithm is then used for matching those descriptors. The proposed method is deletion of certain features after an object has been registered and repetition of successful recognition. The method is expected to recognize all of the registered objects which are shown in an image. A series of tests is done in order to understand the characteristic of the recognizable object and the method capability to do the recognition. The test results show the accuracy of the proposed method is 97% using SURF and 88.7% using SIFT. [ABSTRACT FROM AUTHOR]

Our society is getting more and more technology dependent day by day. Nevertheless, agriculture is imperative for our survival. Rice is one of the primary food grains. It provides sustenance to almost fifty percent of the world population and promotes huge amount of employments. Hence, proper mitigation of rice plant diseases is of paramount importance. A model to detect three rice leaf diseases, namely bacterial leaf blight, brown spot, and leaf smut is proposed in this paper. Backgrounds of the images are removed by saturation threshold while disease affected areas are segmented using hue threshold. Distinctive features from color, shape, and texture domain are extracted from affected areas. These features can robustly describe local and global statistics of such images. Trying a couple of classification algorithms, extreme gradient boosting decision tree ensemble is incorporated in this model for its superior performance. Our model achieves 86.58% accuracy on rice leaf diseases dataset from UCI, which is higher than previous works on the same dataset. Class-wise accuracy of the model is also consistent among the classes. [ABSTRACT FROM AUTHOR]

Currently, convolutional neural networks (CNN) are considered as the most effective tool in image diagnosis and processing techniques. In this paper, we studied and applied the modified SSDLite_MobileNetV2 and proposed a solution to always maintain the boundary of the total memory capacity in the following robust bound and applied on the bridge navigational watch & alarm system (BNWAS). The hardware was designed based on raspberry Pi-3, an embedded single board computer with CPU smartphone level, limited RAM without CUDA GPU. Experimental results showed that the deep learning model on an embedded single board computer brings us high effectiveness in application. [ABSTRACT FROM AUTHOR]

The paper proposed a pornographic image recognition using skin probability and principle component analysis (PCA) on YCbCr color space. The pornographic image recognition is defined as a process to classify the image containing and showing genital elements of human body from any kinds of images. This process is hard to be performed because the images have large variability due to poses, lighting, and background variations. The skin probability and holistic feature, which is extracted by YCbCr skin segmentation and PCA, is employed to handle those variability problems. The function of skin segmentation is to determine skin Region of Interest (ROI) image and skin probability. While the function of PCA is to extract eigenporn of the skin ROIs images and to project the skin ROI vector using the obtained eigenporns to holistic features. The main aim of this research is to optimize the accuracy and false rejection rate of the skin probability and fusion descriptor based recognition system. The experimental result shows that the proposed method can increase the accuracy by about 4.0% and decreases the FPR 20.6% of those of pornographic recognition using fusion descriptors, respectively. In addition, the proposed method is also robust for large size dataset that is shown by giving similar performance to the latest method (Multilayer-Perceptron and Neuro-Fuzzy (MP-NF)). The proposed method also works fast for recognition, which requires 0.12 seconds per image. [ABSTRACT FROM AUTHOR]

One of the major problems that occur in text localization process is the issue of color similarity between text and background image. The limitation of localization algorithms due to high color similarity is highlighted in several research papers. Hence, this research focuses towards the improvement of text localizing capability in high color background image similarity by introducing an adaptive text localization algorithm (ATLAS). ATLAS is an edge-based text localization algorithm that consists of two parts. Text- Background Similarity Index (TBSI) being the first part of ATLAS, measures the similarity index of every text region while the second, Multi Adaptive Threshold (MAT), performs multiple adaptive thresholds calculation using size filtration and degree deviation for locating the possible text region. In this research, ATLAS is verified and compared with other localization techniques based on two parameters, localizing strength and precision. The experiment has been implemented and verified using two types of datasets, generated text color spectrum dataset and Document Analysis and Recognition dataset (ICDAR). The result shows ATLAS has significant improvement on localizing strength and slight improvement on precision compared with other localization algorithms in high color text-background image. [ABSTRACT FROM AUTHOR]

In this paper, we propose an exact image authentication scheme that can, in the best case, detect image tampering with the accuracy of one pixel. This method is based on constructing blocks in the image in such a manner that they intersect with one another in different directions. Such a technique is very useful to identify whether an individual image pixel has been tampered with. Moreover, the tampered region can be well recovered with the embedded recover data. [ABSTRACT FROM AUTHOR]

Some case studies treated by physiotherapists or orthopedists to measure the alignment of the lower extremities during a gait cycle are based on empirical methods of visual observation. This methodology does not guarantee total success, since it depends on the experience of the specialist, what can cause irreversible damage to patients, such as: hip displacement, wear and overload of the joints of a single lower limb. Although, this problem has been addressed in the investigation by means of devices implementation with sensors or methods of processing sequences of images and videos, this topic is still under investigation because the current methods depend on many external elements and data given by an expert in the area. Therefore, this paper proposes a partial solution to this problem by systematizing the experience of a specialist through a computational learning method. [ABSTRACT FROM AUTHOR]

This paper proposes an image processing approach to identify Gram-negative bacteria. Gram-negative bacteria are one of the bacteria that cause lung lobe damage-bacterial samples obtained through smears of the patient's sputum. The first step bacterium should pass the pathogen test process. After that, it bred using Mc Conkey's media. The problem faced is that the process of identifying bacterial objects is still done manually under a fluorescence microscope. The contributions offered from this research are focused on observing bacterial morphology for the operation of selecting shape features. The proposed method is a convolutional neural network with fine-tuning. In the stages of the process, a convolutional neural network of the VGG-16 architecture used dropout, data augmentation, and fine-tuning stages. The main goal of the current research was to determine the method selection is to get a high degree of accuracy. This research uses a total sample of 2520 images from 2 different classes. The amount of data used at each stage of training, testing, and validation is 840 images with dimensions of 256x256 pixels, a resolution of 96 points per inch, and a depth of 24 bits. The accuracy of the results obtained at the training stage is 99.20%. [ABSTRACT FROM AUTHOR]

Removing noise from the image by retaining the details and features of this treated image remains a standing challenge for the researchers in this field. Therefore, this study is carried out to propose and implement a new denoising technique for removing impulse noise from the digital image, using a new way. This technique permits the narrowing of the gap between the original and the restored images, visually and quantitatively by adopting the mathematical concept ''arithmetic progression''. Through this paper, this concept is integrated into the image denoising, due to its ability in modelling the variation of pixels' intensity in the image. The principle of the proposed denoising technique relies on the precision, where it keeps the uncorrupted pixels by using effective noise detection and converts the corrupted pixels by replacing them with other closest pixels from the original image at lower cost and with more simplicity. [ABSTRACT FROM AUTHOR]

Recently, an approach called two-dimensional principal component analysis (2DPCA) has been proposed for smile stages representation and recognition. The essence of 2DPCA is that it computes the eigenvectors of the so-called image covariance matrix without matrix-to-vector conversion so the size of the image covariance matrix are much smaller, easier to evaluate covariance matrix, computation cost is reduced and the performance is also improved than traditional PCA. In an effort to improve and perfect the performance of smile stages recognition, in this paper, we propose efficient Kernel based 2DPCA concepts. The Kernelization of 2DPCA can be benefit to develop the nonlinear structures in the input data. This paper discusses comparison of standard Kernel based 2DPCA and efficient Kernel based 2DPCA for smile stages recognition. The results of experiments show that Kernel based 2DPCA achieve better performance in comparison with the other approaches. While the use of efficient Kernel based 2DPCA can speed up the training procedure of standard Kernel based 2DPCA thus the algorithm can achieve much more computational efficiency and remarkably save the memory consuming compared to the standard Kernel based 2DPCA. [ABSTRACT FROM AUTHOR]

Copyright of Telkomnika is the property of Department of Electrical Engineering, Ahmad Dahlan University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)