Search results for: image gradients

Authors: Hae-Yeoun Lee

Abstract:

Reversible watermarking that not only protects the copyright but also preserve the original quality of the digital content have been intensively studied. In particular, the demand for reversible watermarking has increased. In this paper, we propose a reversible watermarking scheme based on interpolation-error shifting and error pre-compensation. The intensity of a pixel is interpolated from the intensities of neighboring pixels, and the difference histogram between the interpolated and the original intensities is obtained and modified to embed the watermark message. By restoring the difference histogram, the embedded watermark is extracted and the original image is recovered by compensating for the interpolation error. The overflow and underflow are prevented by error pre-compensation. To show the performance of the method, the proposed algorithm is compared with other methods using various test images.

Keywords: Reversible watermarking, High capacity, High quality, Interpolated error shifting, Error pre-compensation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2221

Authors: Nestor A. Garcia Fragoso, Tetyana Baydyk, Ernst Kussul

Abstract:

This article contains information from our investigation in the field of voice recognition. For this purpose, we created a voice database that contains different phrases in two languages, English and Spanish, for men and women. As a classifier, the LIRA (Limited Receptive Area) grayscale neural classifier was selected. The LIRA grayscale neural classifier was developed for image recognition tasks and demonstrated good results. Therefore, we decided to develop a recognition system using this classifier for voice recognition. From a specific set of speakers, we can recognize the speaker’s voice. For this purpose, the system uses spectrograms of the voice signals as input to the system, extracts the characteristics and identifies the speaker. The results are described and analyzed in this article. The classifier can be used for speaker identification in security system or smart buildings for different types of intelligent devices.

Keywords: Extreme learning, LIRA neural classifier, speaker identification, voice recognition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 764

Authors: G. Rezai, P.K.Teng, Z. Mohamed, M.N Shamsudin

Abstract:

Changing in consumers lifestyles and food consumption patterns provide a great opportunity in developing the functional food sector in Malaysia. There is only a little knowledge about whether Malaysian consumers are aware of functional food and if so what image consumers have of this product. The objective of this research is to determine the extent to which selected socioeconomic characteristics and attitudes influence consumers- awareness of functional food. A survey was conducted in the Klang Valley, Malaysia where 439 respondents were interviewed using a structured questionnaire. The result shows that most respondents have a positive attitude towards functional food. For the binary logistic estimation, the results indicate that age, income and other factors such as concern about food safety, subscribing to cooking or health magazines, being a vegetarian and consumers who have been involved in a food production company significantly influence Malaysian consumers- awareness towards functional food.

Keywords: Binary logistic model, functional foods, knowledge and awareness, perception

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 5780

Authors: Dariush Semnani, Javad Yekrang, Hossein Ghayoor

Abstract:

Concerning the measurement of friction properties of textiles and fabrics using Kawabata Evaluation System (KES), whose output is constrained to the surface friction factor of fabric, and no other data would be generated; this research has been conducted to gain information about surface roughness regarding its surface friction factor. To assess roughness properties of light nonwovens, a 3-dimensional model of a surface has been simulated with regular sinuous waves through it as an ideal surface. A new factor was defined, namely Surface Roughness Factor, through comparing roughness properties of simulated surface and real specimens. The relation between the proposed factor and friction factor of specimens has been analyzed by regression, and results showed a meaningful correlation between them. It can be inferred that the new presented factor can be used as an acceptable criterion for evaluating the roughness properties of light nonwoven fabrics.

Keywords: Surface roughness, Nonwoven, Machine vision, Image processing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3092

Authors: Sara A.Yones, Ahmed S. Moussa

Abstract:

Quantitative measurements of tumor in general and tumor volume in particular, become more realistic with the use of Magnetic Resonance imaging, especially when the tumor morphological changes become irregular and difficult to assess by clinical examination. However, tumor volume estimation strongly depends on the image segmentation, which is fuzzy by nature. In this paper a fuzzy approach is presented for tumor volume segmentation based on the fuzzy connectedness algorithm. The fuzzy affinity matrix resulting from segmentation is then used to estimate a fuzzy volume based on a certainty parameter, an Alpha Cut, defined by the user. The proposed method was shown to highly affect treatment decisions. A statistical analysis was performed in this study to validate the results based on a manual method for volume estimation and the importance of using the Alpha Cut is further explained.

Keywords: Alpha Cut, Fuzzy Connectedness, Magnetic Resonance Imaging, Tumor volume estimation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2398

Authors: Ebrahim Taherzadeh, Helmi Z. M. Shafri, Kaveh Shahi

Abstract:

One of the most important tasks in urban remote sensing is the detection of impervious surfaces (IS), such as roofs and roads. However, detection of IS in heterogeneous areas still remains one of the most challenging tasks. In this study, detection of concrete roof using an object-based approach was proposed. A new rule-based classification was developed to detect concrete roof tile. This proposed rule-based classification was applied to WorldView-2 image and results showed that the proposed rule has good potential to predict concrete roof material from WorldView-2 images, with 85% accuracy.

Keywords: Urban remote sensing, impervious surface, Object- Based, Roof Material, Concrete tile, WorldView-2.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3793

Authors: Kahraman Ayyildiz, Stefan Conrad

Abstract:

In this paper we present a system for classifying videos by frequency spectra. Many videos contain activities with repeating movements. Sports videos, home improvement videos, or videos showing mechanical motion are some example areas. Motion of these areas usually repeats with a certain main frequency and several side frequencies. Transforming repeating motion to its frequency domain via FFT reveals these frequencies. Average amplitudes of frequency intervals can be seen as features of cyclic motion. Hence determining these features can help to classify videos with repeating movements. In this paper we explain how to compute frequency spectra for video clips and how to use them for classifying. Our approach utilizes series of image moments as a function. This function again is transformed into its frequency domain.

Keywords: action recognition, frequency feature, motion recognition, repeating movement, video classification

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1884

Authors: Oğuzhan Urhan

Abstract:

In this paper, an improved motion estimation (ME) approach based on weighted constrained one-bit transform is proposed for block-based ME employed in video encoders. Binary ME approaches utilize low bit-depth representation of the original image frames with a Boolean exclusive-OR based hardware efficient matching criterion to decrease computational burden of the ME stage. Weighted constrained one-bit transform (WC‑1BT) based approach improves the performance of conventional C-1BT based ME employing 2-bit depth constraint mask instead of a 1-bit depth mask. In this work, the range of constraint mask is further extended to increase ME performance of WC-1BT approach. Experiments reveal that the proposed method provides better ME accuracy compared existing similar ME methods in the literature.

Keywords: Fast motion estimation, low-complexity motion estimation, video coding.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 854

Authors: Hamid Reza Boveiri

Abstract:

Classification of Persian printed numeral characters has been considered and a proposed system has been introduced. In representation stage, for the first time in Persian optical character recognition, extended moment invariants has been utilized as characters image descriptor. In classification stage, four different classifiers namely minimum mean distance, nearest neighbor rule, multi layer perceptron, and fuzzy min-max neural network has been used, which first and second are traditional nonparametric statistical classifier. Third is a well-known neural network and forth is a kind of fuzzy neural network that is based on utilizing hyperbox fuzzy sets. Set of different experiments has been done and variety of results has been presented. The results showed that extended moment invariants are qualified as features to classify Persian printed numeral characters.

Keywords: Extended moment invariants, optical characterrecognition, Persian numerals classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1919

Authors: Najam Aziz, Nasru Minallah, Ahmad Junaid, Kashaf Gul

Abstract:

Landcover classification using automated classification techniques, while employing remotely sensed multi-spectral imagery, is one of the promising areas of research. Different land conditions at different time are captured through satellite and monitored by applying different classification algorithms in specific environment. In this paper, a SPOT-5 image provided by SUPARCO has been studied and classified in Environment for Visual Interpretation (ENVI), a tool widely used in remote sensing. Then, Artificial Neural Network (ANN) classification technique is used to detect the land cover changes in Abbottabad district. Obtained results are compared with a pixel based Distance classifier. The results show that ANN gives the better overall accuracy of 99.20% and Kappa coefficient value of 0.98 over the Mahalanobis Distance Classifier.

Keywords: Landcover classification, artificial neural network, remote sensing, SPOT-5.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1606

Authors: Ahsan Bin Tufail, Ali Abidi, Adil Masood Siddiqui, Muhammad Shahzad Younis

Abstract:

An early and accurate detection of Alzheimer's disease (AD) is an important stage in the treatment of individuals suffering from AD. We present an approach based on the use of structural magnetic resonance imaging (sMRI) phase images to distinguish between normal controls (NC), mild cognitive impairment (MCI) and AD patients with clinical dementia rating (CDR) of 1. Independent component analysis (ICA) technique is used for extracting useful features which form the inputs to the support vector machines (SVM), K nearest neighbour (kNN) and multilayer artificial neural network (ANN) classifiers to discriminate between the three classes. The obtained results are encouraging in terms of classification accuracy and effectively ascertain the usefulness of phase images for the classification of different stages of Alzheimer-s disease.

Keywords: Biomedical image processing, classification algorithms, feature extraction, statistical learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2765

Authors: M. Ovinis, D. Kerr, K. Bouazza-Marouf, M. Vloeberghs

Abstract:

The use of machine vision to inspect the outcome of surgical tasks is investigated, with the aim of incorporating this approach in robotic surgery systems. Machine vision is a non-contact form of inspection i.e. no part of the vision system is in direct contact with the patient, and is therefore well suited for surgery where sterility is an important consideration,. As a proof-of-concept, three primary surgical tasks for a common neurosurgical procedure were inspected using machine vision. Experiments were performed on cadaveric pig heads to simulate the two possible outcomes i.e. satisfactory or unsatisfactory, for tasks involved in making a burr hole, namely incision, retraction, and drilling. We identify low level image features to distinguish the two outcomes, as well as report on results that validate our proposed approach. The potential of using machine vision in a surgical environment, and the challenges that must be addressed, are identified and discussed.

Keywords: Visual inspection, machine vision, robotic surgery.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1800

Authors: Siti Norul Huda Sheikh Abdullah, Farshid Pirahan Siah, Nor Hanisah Haji Zainal Abidin, Shahnorbanun Sahran

Abstract:

The objective of this paper is to propose an adaptive multi threshold for image segmentation precisely in object detection. Due to the different types of license plates being used, the requirement of an automatic LPR is rather different for each country. The proposed technique is applied on Malaysian LPR application. It is based on Multi Layer Perceptron trained by back propagation. The proposed adaptive threshold is introduced to find the optimum threshold values. The technique relies on the peak value from the graph of the number object versus specific range of threshold values. The proposed approach has improved the overall performance compared to current optimal threshold techniques. Further improvement on this method is in progress to accommodate real time system specification.

Keywords: Multi-threshold approach, license plate recognition system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2523

Authors: J. Silvestre, L. Almeida, R. Marau, P. Pedreiras

Abstract:

Currently, there are many local area industrial networks that can give guaranteed bandwidth to synchronous traffic, particularly providing CBR channels (Constant Bit Rate), which allow improved bandwidth management. Some of such networks operate over Ethernet, delivering channels with enough capacity, specially with compressors, to integrate multimedia traffic in industrial monitoring and image processing applications with many sources. In these industrial environments where a low latency is an essential requirement, JPEG is an adequate compressing technique but it generates VBR traffic (Variable Bit Rate). Transmitting VBR traffic in CBR channels is inefficient and current solutions to this problem significantly increase the latency or further degrade the quality. In this paper an R(q) model is used which allows on-line calculation of the JPEG quantification factor. We obtained increased quality, a lower requirement for the CBR channel with reduced number of discarded frames along with better use of the channel bandwidth.

Keywords: Industrial Networks, Multimedia.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1594

Authors: D. Sumathi, P. Poongodi

Abstract:

Cloud Computing refers to applications delivered as services over the internet, and the datacenters that provide those services with hardware and systems software. These were earlier referred to as Software as a Service (SaaS). Scheduling is justified by job components (called tasks), lack of information. In fact, in a large fraction of jobs from machine learning, bio-computing, and image processing domains, it is possible to estimate the maximum time required for a task in the job. This study focuses on Trust based scheduling to improve cloud security by modifying Heterogeneous Earliest Finish Time (HEFT) algorithm. It also proposes TR-HEFT (Trust Reputation HEFT) which is then compared to Dynamic Load Scheduling.

Keywords: Software as a Service (SaaS), Trust, Heterogeneous Earliest Finish Time (HEFT) algorithm, Dynamic Load Scheduling.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2195

Authors: Nareenad Panbun

Abstract:

The objective is to study the knowledge utilization from the participants of the announcer training program by National Broadcasting and Telecommunications Commission (NBTC). This study is a quantitative research based on surveys and self-answering questionnaires. The population of this study is 100 participants randomly chosen by non-probability sampling method. The results have shown that most of the participants were satisfied with the topics of general knowledge about the broadcasting and television business for 37 people representing 37%, followed by the topics of broadcasting techniques. The legal issues, consumer rights, television business ethics, and credibility of the media are, in addition to the media's role and responsibilities in society, the use of language for successful communication. Therefore, the communication language skills are the most important for all of the trainees and will also build up the image of the broadcasting center.

Keywords: Announcer training program, participant, requirements announced, theory of utilization.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 751

Authors: A. Mohammadi Torkashvand, N. Haghighat

Abstract:

Some methodologies were compared in providing erosion maps of surface, rill and gully and erosion features, in research which took place in the Varamin sub-basin, north-east Tehran, Iran. A photomorphic unit map was produced from processed satellite images, and four other maps were prepared by the integration of different data layers, including slope, plant cover, geology, land use, rocks erodibility and land units. Comparison of ground truth maps of erosion types and working unit maps indicated that the integration of land use, land units and rocks erodibility layers with satellite image photomorphic units maps provide the best methods in producing erosion types maps.

Keywords: Erosion Features, Geographic Information System, Remote Sensing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1780

Authors: Erik Lilienblum, Robert Niese, Bernd Michaelis

Abstract:

This paper proposes a novel stereo vision technique for top view book scanners which provide us with dense 3d point clouds of page surfaces. This is a precondition to dewarp bound volumes independent of 2d information on the page. Our method is based on algorithms, which normally require the projection of pattern sequences with structured light. We use image sequences of the moving stripe lighting of the top view scanner instead of an additional light projection. Thus the stereo vision setup is simplified without losing measurement accuracy. Furthermore we improve a surface model dewarping method through introducing a difference vector based on real measurements. Although our proposed method is hardly expensive neither in calculation time nor in hardware requirements we present good dewarping results even for difficult examples.

Keywords: stereo vision, 3d surface reconstruction, dewarpingdocuments, book scanner

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1588

Authors: Eva Štěpánková

Abstract:

Environmental management implementation is presently one of the ways of organization success and value improvement. Increasing an organization motivation to environmental measures introduction is caused primarily by the rising pressure of the society that generates various incentives to endeavor for the environmental performance improvement. The aim of the paper is to identify and characterize the key incentives and expectations leading organizations to the environmental management implementation. The author focuses on five businesses of different size and field, operating in the Czech Republic. The qualitative approach and grounded theory procedure are used in research. The results point out that the significant incentives for environmental management implementation represent primarily demands of customers, the opportunity to declare the environmental commitment and image improvement. The researched enterprises less commonly expect the economical contribution, competitive advantage increase or export rate improvement. The results show that marketing contributions are primarily expected from the environmental management implementation.

Keywords: Environmental management, environmental management systems, ISO 14001.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2541

Authors: Udeni Jayasinghe, Anuja Dharmaratne

Abstract:

Each year many people are reported missing in most of the countries in the world owing to various reasons. Arrangements have to be made to find these people after some time. So the investigating agencies are compelled to make out these people by using manpower. But in many cases, the investigations carried out to find out an absconding for a long time may not be successful. At a time like that it may be difficult to identify these people by examining their old photographs, because their facial appearance might have changed mainly due to the natural aging process. On some occasions in forensic medicine if a dead body is found, investigations should be held to make sure that this corpse belongs to the same person disappeared some time ago. With the passage of time the face of the person might have changed and there should be a mechanism to reveal the person-s identity. In order to make this process easy, we must guess and decide as to how he will look like by now. To address this problem this paper presents a way of synthesizing a facial image with the aging effects.

Keywords: Cranio-facial growth model, eigenfaces, eigenvectors, Face Anthropometry.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1741

Authors: K. A. Navas, S. Archana Thampy, M. Sasikumar

Abstract:

Medical image data hiding has strict constrains such as high imperceptibility, high capacity and high robustness. Achieving these three requirements simultaneously is highly cumbersome. Some works have been reported in the literature on data hiding, watermarking and stegnography which are suitable for telemedicine applications. None is reliable in all aspects. Electronic Patient Report (EPR) data hiding for telemedicine demand it blind and reversible. This paper proposes a novel approach to blind reversible data hiding based on integer wavelet transform. Experimental results shows that this scheme outperforms the prior arts in terms of zero BER (Bit Error Rate), higher PSNR (Peak Signal to Noise Ratio), and large EPR data embedding capacity with WPSNR (Weighted Peak Signal to Noise Ratio) around 53 dB, compared with the existing reversible data hiding schemes.

Keywords: Biomedical imaging, Data security, Datacommunication, Teleconferencing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2754

Authors: Mario Mastriani, Marcelo Naiouf

Abstract:

This paper introduces a Quantum Correlation Matrix Memory (QCMM) and Enhanced QCMM (EQCMM), which are useful to work with quantum memories. A version of classical Gram-Schmidt orthogonalisation process in Dirac notation (called Quantum Orthogonalisation Process: QOP) is presented to convert a non-orthonormal quantum basis, i.e., a set of non-orthonormal quantum vectors (called qudits) to an orthonormal quantum basis, i.e., a set of orthonormal quantum qudits. This work shows that it is possible to improve the performance of QCMM thanks QOP algorithm. Besides, the EQCMM algorithm has a lot of additional fields of applications, e.g.: Steganography, as a replacement Hopfield Networks, Bilevel image processing, etc. Finally, it is important to mention that the EQCMM is an extremely easy to implement in any firmware.

Keywords: Quantum Algebra, correlation matrix memory, Dirac notation, orthogonalisation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1719

Authors: Sana Ktata, Kaïs Ouni, Noureddine Ellouze

Abstract:

Electrocardiogram (ECG) data compression algorithm is needed that will reduce the amount of data to be transmitted, stored and analyzed, but without losing the clinical information content. A wavelet ECG data codec based on the Set Partitioning In Hierarchical Trees (SPIHT) compression algorithm is proposed in this paper. The SPIHT algorithm has achieved notable success in still image coding. We modified the algorithm for the one-dimensional (1-D) case and applied it to compression of ECG data. By this compression method, small percent root mean square difference (PRD) and high compression ratio with low implementation complexity are achieved. Experiments on selected records from the MIT-BIH arrhythmia database revealed that the proposed codec is significantly more efficient in compression and in computation than previously proposed ECG compression schemes. Compression ratios of up to 48:1 for ECG signals lead to acceptable results for visual inspection.

Keywords: Discrete Wavelet Transform, ECG compression, SPIHT.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2131

Authors: Tin Hninn Hninn Maung

Abstract:

This paper introduces a hand gesture recognition system to recognize real time gesture in unstrained environments. Efforts should be made to adapt computers to our natural means of communication: Speech and body language. A simple and fast algorithm using orientation histograms will be developed. It will recognize a subset of MAL static hand gestures. A pattern recognition system will be using a transforrn that converts an image into a feature vector, which will be compared with the feature vectors of a training set of gestures. The final system will be Perceptron implementation in MATLAB. This paper includes experiments of 33 hand postures and discusses the results. Experiments shows that the system can achieve a 90% recognition average rate and is suitable for real time applications.

Keywords: Hand gesture recognition, Orientation Histogram, Myanmar Alphabet Language, Perceptronnetwork, MATLAB.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4697

Authors: Sello Mokwena, Monyepao Thabang

Abstract:

In this study, we propose a comprehensive approach to address the challenges in person re-identification models. By combining a centroid tracking algorithm with a Siamese convolutional neural network model, our method excels in detecting, tracking, and capturing robust person features across non-overlapping camera views. The algorithm efficiently identifies individuals in the camera network, while the neural network extracts fine-grained global features for precise cross-image comparisons. The approach's effectiveness is further accentuated by leveraging the camera network topology for guidance. Our empirical analysis of benchmark datasets highlights its competitive performance, particularly evident when background subtraction techniques are selectively applied, underscoring its potential in advancing person re-identification techniques.

Keywords: Camera network, convolutional neural network topology, person tracking, person re-identification, Siamese.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 81

Authors: N. Valliammal, S.N. Geethalakshmi

Abstract:

This paper describes an optimal approach for feature subset selection to classify the leaves based on Genetic Algorithm (GA) and Kernel Based Principle Component Analysis (KPCA). Due to high complexity in the selection of the optimal features, the classification has become a critical task to analyse the leaf image data. Initially the shape, texture and colour features are extracted from the leaf images. These extracted features are optimized through the separate functioning of GA and KPCA. This approach performs an intersection operation over the subsets obtained from the optimization process. Finally, the most common matching subset is forwarded to train the Support Vector Machine (SVM). Our experimental results successfully prove that the application of GA and KPCA for feature subset selection using SVM as a classifier is computationally effective and improves the accuracy of the classifier.

Keywords: Optimization, Feature extraction, Feature subset, Classification, GA, KPCA, SVM and Computation

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2241

Authors: Jaeyoung Lee

Abstract:

Autonomous driving systems require high reliability to provide people with a safe and comfortable driving experience. However, despite the development of a number of vehicle sensors, it is difficult to always provide high perceived performance in driving environments that vary from time to season. The image segmentation method using deep learning, which has recently evolved rapidly, provides high recognition performance in various road environments stably. However, since the system controls a vehicle in real time, a highly complex deep learning network cannot be used due to time and memory constraints. Moreover, efficient networks are optimized for GPU environments, which degrade performance in embedded processor environments equipped simple hardware accelerators. In this paper, a semantic segmentation network, matrix multiplication accelerator network (MMANet), optimized for matrix multiplication accelerator (MMA) on Texas instrument digital signal processors (TI DSP) is proposed to improve the recognition performance of autonomous driving system. The proposed method is designed to maximize the number of layers that can be performed in a limited time to provide reliable driving environment information in real time. First, the number of channels in the activation map is fixed to fit the structure of MMA. By increasing the number of parallel branches, the lack of information caused by fixing the number of channels is resolved. Second, an efficient convolution is selected depending on the size of the activation. Since MMA is a fixed, it may be more efficient for normal convolution than depthwise separable convolution depending on memory access overhead. Thus, a convolution type is decided according to output stride to increase network depth. In addition, memory access time is minimized by processing operations only in L3 cache. Lastly, reliable contexts are extracted using the extended atrous spatial pyramid pooling (ASPP). The suggested method gets stable features from an extended path by increasing the kernel size and accessing consecutive data. In addition, it consists of two ASPPs to obtain high quality contexts using the restored shape without global average pooling paths since the layer uses MMA as a simple adder. To verify the proposed method, an experiment is conducted using perfsim, a timing simulator, and the Cityscapes validation sets. The proposed network can process an image with 640 x 480 resolution for 6.67 ms, so six cameras can be used to identify the surroundings of the vehicle as 20 frame per second (FPS). In addition, it achieves 73.1% mean intersection over union (mIoU) which is the highest recognition rate among embedded networks on the Cityscapes validation set.

Keywords: Edge network, embedded network, MMA, matrix multiplication accelerator and semantic segmentation network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 467

Authors: Ali Abdrhman Ukasha

Abstract:

This paper presented two new efficient algorithms for contour approximation. The proposed algorithm is compared with Ramer (good quality), Triangle (faster) and Trapezoid (fastest) in this work; which are briefly described. Cartesian co-ordinates of an input contour are processed in such a manner that finally contours is presented by a set of selected vertices of the edge of the contour. In the paper the main idea of the analyzed procedures for contour compression is performed. For comparison, the mean square error and signal-to-noise ratio criterions are used. Computational time of analyzed methods is estimated depending on a number of numerical operations. Experimental results are obtained both in terms of image quality, compression ratios, and speed. The main advantages of the analyzed algorithm is small numbers of the arithmetic operations compared to the existing algorithms.

Keywords: Polygonal approximation, Ramer, Triangle and Trapezoid methods.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1805

Authors: Hirofumi Taki, Takuya Sakamoto, Makoto Yamakawa, Tsuyoshi Shiina, Toru Sato

Abstract:

For the improvement of the ability in detecting small calcifications using Ultrasonography (US) we propose a novel indicator of calcifications in an ultrasound B-mode image without decrease in frame rate. Since the waveform of an ultrasound pulse changes at a calcification position, the decorrelation of adjacent scan lines occurs behind a calcification. Therefore, we employ the decorrelation of adjacent scan lines as an indicator of a calcification. The proposed indicator depicted wires 0.05 mm in diameter at 2 cm depth with a sensitivity of 86.7% and a specificity of 100%, which were hardly detected in ultrasound B-mode images. This study shows the potential of the proposed indicator to approximate the detectable calcification size using an US device to that of an X-ray imager, implying the possibility that an US device will become a convenient, safe, and principal clinical tool for the screening of breast cancer.

Keywords: Ultrasonography, Calcification, Decorrelation, Forward scattered wave

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1452

Authors: Sumeet S. Nisale, Chandan J. Bharambe, Vidya N.More

Abstract:

We propose our genuine research of geometric moments which detects the mineral inadequacy in the frail groundnut plant. This plant is prone to many deficiencies as a result of the variance in the soil nutrients. By analyzing the leaves of the plant, we detect the visual symptoms that are not recognizable to the naked eyes. We have collected about 160 samples of leaves from the nearby fields. The images have been taken by keeping every leaf into a black box to avoid the external interference. For the first time, it has been possible to provide the farmer with the stages of deficiencies. This paper has applied the algorithms successfully to many other plants like Lady-s finger, Green Bean, Lablab Bean, Chilli and Tomato. But we submit the results of the groundnut predominantly. The accuracy of our algorithm and method is almost 93%. This will again pioneer a kind of green revolution in the field of agriculture and will be a boon to that field.

Keywords: Component image, geometric moments, average intensity, average affected area, black box

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2132