Abstract: This paper reviews the different concepts that have been investigated concerning the mechanization of fruit thinning as well as multiple working principles and solutions that have been developed for feature extraction of horticultural products, both in the field and industrial environments. The research should be committed towards selective methods, which inevitably need to incorporate some kinds of sensor technology. Computer vision often comes out as an obvious solution for unstructured detection problems, although leaves despite the chosen point of view frequently occlude fruits. Further research on non-traditional sensors that are capable of object differentiation is needed. Ultrasonic and Near Infrared (NIR) technologies have been investigated for applications related to horticultural produce and show a potential to satisfy this need while simultaneously providing spatial information as time of flight sensors. Light Detection and Ranging (LIDAR) technology also shows a huge potential but it implies much greater costs and the related equipment is usually much larger, making it less suitable for portable devices, which may serve a purpose on smaller unstructured orchards. Portable devices may serve a purpose on these types of orchards. In what concerns sensor methods, on-tree fruit detection, major challenge is to overcome the problem of fruits’ occlusion by leaves and branches. Hence, nontraditional sensors capable of providing some type of differentiation should be investigated.
Abstract: Image segmentation is a method to extract regions of interest from an image. It remains a fundamental problem in computer vision. The increasing diversity and the complexity of segmentation algorithms have led us firstly, to make a review and classify segmentation techniques, secondly to identify the most used measures of segmentation performance and thirdly, discuss deeply on segmentation philosophy in order to help the choice of adequate segmentation techniques for some applications. To justify the relevance of our analysis, recent algorithms of segmentation are presented through the proposed classification.
Abstract: Computacional recognition of sign languages aims to
allow a greater social and digital inclusion of deaf people through
interpretation of their language by computer. This article presents
a model of recognition of two of global parameters from sign
languages; hand configurations and hand movements. Hand motion
is captured through an infrared technology and its joints are built
into a virtual three-dimensional space. A Multilayer Perceptron
Neural Network (MLP) was used to classify hand configurations and
Dynamic Time Warping (DWT) recognizes hand motion. Beyond
of the method of sign recognition, we provide a dataset of
hand configurations and motion capture built with help of fluent
professionals in sign languages. Despite this technology can be
used to translate any sign from any signs dictionary, Brazilian
Sign Language (Libras) was used as case study. Finally, the model
presented in this paper achieved a recognition rate of 80.4%.
Abstract: Ultrasonic infrared nondestructive testing is a kind of testing method with high speed, accuracy and localization. However, there are still some problems, such as the detection requires manual real-time field judgment, the methods of result storage and viewing are still primitive. An intelligent non-destructive detection system based on embedded linux is put forward in this paper. The hardware part of the detection system is based on the ARM (Advanced Reduced Instruction Set Computer Machine) core and an embedded linux system is built to realize image processing and defect detection of thermal images. The CLAHE algorithm and the Butterworth filter are used to process the thermal image, and then the boa server and CGI (Common Gateway Interface) technology are used to transmit the test results to the display terminal through the network for real-time monitoring and remote monitoring. The system also liberates labor and eliminates the obstacle of manual judgment. According to the experiment result, the system provides a convenient and quick solution for industrial non-destructive testing.
Abstract: The research follows a systematic approach to optimize the parameters for parts machined by turning and milling processes. The quality characteristic chosen is surface roughness since the surface finish plays an important role for parts that require surface contact. A tapered cylindrical surface is designed as a test specimen for the research. The material chosen for machining is aluminum alloy 6061 due to its wide variety of industrial and engineering applications. HAAS VF-2 TR computer numerical control (CNC) vertical machining center is used for milling and HAAS ST-20 CNC machine is used for turning in this research. Taguchi analysis is used to optimize the surface roughness of the machined parts. The L9 Orthogonal Array is designed for four controllable factors with three different levels each, resulting in 18 experimental runs. Signal to Noise (S/N) Ratio is calculated for achieving the specific target value of 75 ± 15 µin. The controllable parameters chosen for turning process are feed rate, depth of cut, coolant flow and finish cut and for milling process are feed rate, spindle speed, step over and coolant flow. The uncontrollable factors are tool geometry for turning process and tool material for milling process. Hypothesis testing is conducted to study the significance of different uncontrollable factors on the surface roughnesses. The optimal parameter settings were identified from the Taguchi analysis and the process capability Cp and the process capability index Cpk were improved from 1.76 and 0.02 to 3.70 and 2.10 respectively for turning process and from 0.87 and 0.19 to 3.85 and 2.70 respectively for the milling process. The surface roughnesses were improved from 60.17 µin to 68.50 µin, reducing the defect rate from 52.39% to 0% for the turning process and from 93.18 µin to 79.49 µin, reducing the defect rate from 71.23% to 0% for the milling process. The purpose of this study is to efficiently utilize the Taguchi design analysis to improve the surface roughness.
Abstract: In this paper, we conduct a systematic survey of urban communities in Lithuania to evaluate their potential to co-create collective intelligence or “civic intelligence” applying Digital Co-creation Index methodology that includes different socio-technological indicators. Civic intelligence is a form of collective intelligence that refers to the group’s capacity to perceive societal problems and to address them effectively. The research focuses on evaluation of diverse organizational designs that increase efficient collective performance. The current scientific project advanced the state of the art by evaluating the basic preconditions in the urban communities through which the collective intelligence is being co-created under the systemic manner. The research subject is the “bottom up” digital enabled urban platforms, initiated by Lithuanian public organizations, civic movements or business entities. The web-based monitoring results obtained by applying a social indices calculation methodology and Pearson correlation analysis provided the information about the potential and limits of the urban communities and what possible changes need to be implemented to overcome the limitations.
Abstract: Studies estimate that there will be 266,120 new cases
of invasive breast cancer and 40,920 breast cancer induced deaths
in the year of 2018 alone. Despite the pervasiveness of this
affliction, the current process to obtain an accurate breast cancer
prognosis is tedious and time consuming. It usually requires a
trained pathologist to manually examine histopathological images and
identify the features that characterize various cancer severity levels.
We propose MITOS-RCNN: a region based convolutional neural
network (RCNN) geared for small object detection to accurately
grade one of the three factors that characterize tumor belligerence
described by the Nottingham Grading System: mitotic count. Other
computational approaches to mitotic figure counting and detection
do not demonstrate ample recall or precision to be clinically viable.
Our models outperformed all previous participants in the ICPR 2012
challenge, the AMIDA 2013 challenge and the MITOS-ATYPIA-14
challenge along with recently published works. Our model achieved
an F- measure score of 0.955, a 6.11% improvement in accuracy from
the most accurate of the previously proposed models.
Abstract: This paper describes the design and application of a short in-class project conducted in Algoma University’s Human-Computer Interaction (HCI) course taught at the Bachelor of Computer Science. The project was based on the Maker Movement (people using and reusing electronic components and everyday materials to tinker with technology and make interactive applications), where students applied low-cost and easy-to-use electronic components, the Arduino Uno microcontroller board, software tools, and everyday objects. Students collaborated in small teams by completing hands-on activities with them, making an interactive walking cane for blind people. At the end of the course, students filled out a Technology Acceptance Model version 2 (TAM2) questionnaire where they evaluated microcontroller boards’ applications in HCI classes. We also asked them about applying the Maker Movement in HCI classes. Results showed overall students’ positive opinions and response about using microcontroller boards in HCI classes. We strongly suggest that every HCI course should include practical activities related to tinkering with technology such as applying microcontroller boards, where students actively and constructively participate in teams for achieving learning objectives.
Abstract: In this paper, an interactive in-car interface called HoloDash is presented. It is intended to provide information and infotainment in both autonomous vehicles and ‘connected cars’, vehicles equipped with Internet access via cellular services. The research focuses on the development of interactive avatars for this system and its gesture-based control system. This is a case study for the development of a possible human-centred means of presenting a connected or autonomous vehicle’s On-Board Diagnostics through a projected ‘holographic’ infotainment system. This system is termed a Holographic Human Vehicle Interface (HHIV), as it utilises a dashboard projection unit and gesture detection. The research also examines the suitability for gestures in an automotive environment, given that it might be used in both driver-controlled and driverless vehicles. Using Human Centred Design methods, questions were posed to test subjects and preferences discovered in terms of the gesture interface and the user experience for passengers within the vehicle. These affirm the benefits of this mode of visual communication for both connected and driverless cars.
Abstract: Lack of motivation and interest is a serious obstacle to students’ learning computing skills. A need exists for a knowledge base on effective pedagogy and curricula to teach computer programming. This paper presents results from research evaluating a six-year project designed to teach complex concepts in computer programming collaboratively, while supporting students to continue developing their computer thinking and related coding skills individually. Utilizing a quasi-experimental, mixed methods design, the pedagogical approaches and methods were assessed in two contrasting groups of students with different socioeconomic status, gender, and age composition. Analyses of quantitative data from Likert-scale surveys and an evaluation rubric, combined with qualitative data from reflective writing exercises and semi-structured interviews yielded convincing evidence of the project’s success at both teaching and inspiring students.
Abstract: Digital multimedia contents such as image, video, and audio can be tampered easily due to the availability of powerful editing softwares. Multimedia forensics is devoted to analyze these contents by using various digital forensic techniques in order to validate their authenticity. Digital image forensics is dedicated to investigate the reliability of digital images by analyzing the integrity of data and by reconstructing the historical information of an image related to its acquisition phase. In this paper, a survey is carried out on the forgery detection by considering the most recent and promising digital image forensic techniques.
Abstract: Image registration is an important topic for many imaging systems and computer vision applications. The standard image registration techniques such as Mutual information/ Normalized mutual information -based methods have a limited performance because they do not consider the spatial information or the relationships between the neighbouring pixels or voxels. In addition, the amount of image noise may significantly affect the registration accuracy. Therefore, this paper proposes an efficient method that explicitly considers the relationships between the adjacent pixels, where the gradient information of the reference and scene images is extracted first, and then the cosine similarity of the extracted gradient information is computed and used to improve the accuracy of the standard normalized mutual information measure. Our experimental results on different data types (i.e. CT, MRI and thermal images) show that the proposed method outperforms a number of image registration techniques in terms of the accuracy.
Abstract: The use of Information and Communication Technology (ICT) in tertiary institutions by lecturers and students has become a necessity for the enhancement of quality teaching and learning. This study examined availability, accessibility and utilization of ICT in Teaching-Learning Islamic Studies in Colleges of Education, North-East, Nigeria. The study adopted multi-stage sampling technique, in which, five out of the eleven Colleges of Education (both Federal and State owned) were purposively selected for the study. Primary data was drawn from the respondents by the use of questionnaire, interviews and observations. The results of the study, generally, indicate that the availability and accessibility to ICT facilities in Colleges of Education in North-East, Nigeria, especially in teaching/learning delivery of Islamic studies were relatively inadequate and rare to lecturers and students. The study further reveals that the respondents’ level of utilization of ICT is low and only few computer packages and internet services were involved in the ICT utilization, which is yet to reach the real expected situation of the globalization and advancement in the application of ICT if compared to other parts of the world, as far as the teaching and learning of Islamic studies is concerned. Observations and conclusion were drawn from the findings and finally, recommendations on how to improve on ICT availability, accessibility and utilization in teaching/ learning were suggested.
Abstract: This case study explores the impact of two major computer software programs Learn to Speak English and Learn English Spelling and Pronunciation, and some Internet search engines such as Google on mending the decoding and spelling deficiency of Simon X, a dyslexic student. The improvement in decoding and spelling may result in better reading comprehension and composition writing. Some computer programs and Internet materials can help regain the missing awareness and consequently restore his self-confidence and self-esteem. In addition, this study provides a systematic plan comprising a set of activities (four computer programs and Internet materials) which address the problem from the lowest to the highest levels of phoneme and phonological awareness. Four methods of data collection (accounts, observations, published tests, and interviews) create the triangulation to validly and reliably collect data before the plan, during the plan, and after the plan. The data collected are analyzed quantitatively and qualitatively. Sometimes the analysis is either quantitative or qualitative, and some other times a combination of both. Tables and figures are utilized to provide a clear and uncomplicated illustration of some data. The improvement in the decoding, spelling, reading comprehension, and composition writing skills that occurred is proved through the use of authentic materials performed by the student under study. Such materials are a comparison between two sample passages written by the learner before and after the plan, a genuine computer chat conversation, and the scores of the academic year that followed the execution of the plan. Based on these results, the researcher recommends further studies on other Lebanese dyslexic learners using the computer to mend their language problem in order to design and make a most reliable software program that can address this disability more efficiently and successfully.
Abstract: In this paper, we propose an optimized brain computer
interface (BCI) system for unspoken speech recognition, based on
the fact that the constructions of unspoken words rely strongly on the
Wernicke area, situated in the temporal lobe. Our BCI system has four
modules: (i) the EEG Acquisition module based on a non-invasive
headset with 14 electrodes; (ii) the Preprocessing module to remove
noise and artifacts, using the Common Average Reference method;
(iii) the Features Extraction module, using Wavelet Packet Transform
(WPT); (iv) the Classification module based on a one-hidden layer
artificial neural network. The present study consists of comparing
the recognition accuracy of 5 Arabic words, when using all the
headset electrodes or only the 4 electrodes situated near the Wernicke
area, as well as the selection effect of the subbands produced by
the WPT module. After applying the articial neural network on the
produced database, we obtain, on the test dataset, an accuracy of
83.4% with all the electrodes and all the subbands of 8 levels of the
WPT decomposition. However, by using only the 4 electrodes near
Wernicke Area and the 6 middle subbands of the WPT, we obtain
a high reduction of the dataset size, equal to approximately 19% of
the total dataset, with 67.5% of accuracy rate. This reduction appears
particularly important to improve the design of a low cost and simple
to use BCI, trained for several words.
Abstract: Knee orthotics play an important role in aiding in the recovery of those with knee injuries, especially athletes. However, structural knee orthotics is often very expensive, ranging between $300 and $800. The primary reason for this project was to answer the question: can 3D printed orthotics represent a viable and cost-effective alternative to present structural knee orthotics? The primary objective for this research project was to design a knee orthotic for athletes with knee injuries for a low-cost under $100 and evaluate its effectiveness. The initial design for the orthotic was done in SolidWorks, a computer-aided design (CAD) software available at Loyola Marymount University. After this design was completed, finite element analysis (FEA) was utilized to understand how normal stresses placed upon the knee affected the orthotic. The knee orthotic was then adjusted and redesigned to meet a specified factor-of-safety of 3.25 based on the data gathered during FEA and literature sources. Once the FEA was completed and the orthotic was redesigned based from the data gathered, the next step was to move on to 3D-printing the first design of the knee brace. Subsequently, physical therapy movement trials were used to evaluate physical performance. Using the data from these movement trials, the CAD design of the brace was refined to accommodate the design requirements. The final goal of this research means to explore the possibility of replacing high-cost, outsourced knee orthotics with a readily available low-cost alternative.
Abstract: This paper presents a deep-learning mechanism for classifying computer generated images and photographic images. The proposed method accounts for a convolutional layer capable of automatically learning correlation between neighbouring pixels. In the current form, Convolutional Neural Network (CNN) will learn features based on an image's content instead of the structural features of the image. The layer is particularly designed to subdue an image's content and robustly learn the sensor pattern noise features (usually inherited from image processing in a camera) as well as the statistical properties of images. The paper was assessed on latest natural and computer generated images, and it was concluded that it performs better than the current state of the art methods.
Abstract: We here describe the theoretical and philosophical understanding of a long term use and development of algorithmic computer-based tools applied to music composition. The findings of our research lead us to interrogate some specific processes and systems of communication engaged in the discovery of specific cultural artworks: artistic creation in the sono-musical domain. Our hypothesis is that the patterns of auditory learning cannot be only understood in terms of social transmission but would gain to be questioned in the way they rely on various ranges of acoustic stimuli modes of consciousness and how the different types of memories engaged in the percept-action expressive systems of our cultural communities also relies on these shadowy conscious entities we named “Reduced Descriptive Structures”.
Abstract: Technological and sociological developments in the automotive sector are shifting the focus of design towards developing a better understanding of driver needs, desires and emotions. Human centred design methods are being more frequently applied to automotive research, including the use of systems to detect human emotions in real-time. One method for a non-contact measurement of emotion with low intrusiveness is Facial-Expression Analysis (FEA). This paper describes a research study investigating emotional responses of 22 participants in a naturalistic driving environment by applying a multi-method approach. The research explored the possibility to investigate emotional responses and their frequencies during naturalistic driving through real-time FEA. Observational analysis was conducted to assign causes to the collected emotional responses. In total, 730 emotional responses were measured in the collective study time of 440 minutes. Causes were assigned to 92% of the measured emotional responses. This research establishes and validates a methodology for the study of emotions and their causes in the driving environment through which systems and factors causing positive and negative emotional effects can be identified.
Abstract: Human motion recognition has been extensively increased in recent years due to its importance in a wide range of applications, such as human-computer interaction, intelligent surveillance, augmented reality, content-based video compression and retrieval, etc. However, it is still regarded as a challenging task especially in realistic scenarios. It can be seen as a general machine learning problem which requires an effective human motion representation and an efficient learning method. In this work, we introduce a descriptor based on Laban Movement Analysis technique, a formal and universal language for human movement, to capture both quantitative and qualitative aspects of movement. We use Discrete Hidden Markov Model (DHMM) for training and classification motions. We improve the classification algorithm by proposing two DHMMs for each motion class to process the motion sequence in two different directions, forward and backward. Such modification allows avoiding the misclassification that can happen when recognizing similar motions. Two experiments are conducted. In the first one, we evaluate our method on a public dataset, the Microsoft Research Cambridge-12 Kinect gesture data set (MSRC-12) which is a widely used dataset for evaluating action/gesture recognition methods. In the second experiment, we build a dataset composed of 10 gestures(Introduce yourself, waving, Dance, move, turn left, turn right, stop, sit down, increase velocity, decrease velocity) performed by 20 persons. The evaluation of the system includes testing the efficiency of our descriptor vector based on LMA with basic DHMM method and comparing the recognition results of the modified DHMM with the original one. Experiment results demonstrate that our method outperforms most of existing methods that used the MSRC-12 dataset, and a near perfect classification rate in our dataset.