The increased in number and size of larvae and juvenile growth are estimated based on manual approach in fishing hatcheries. There is a high demand for computer assisted software solution for aquaculture research in early detection and recognition of fish population. There exist several companies who have introduced fish detection technologies into the market. Although able to count the number of larvae with a high accuracy rate, the fish counter software's may encounter difficulties when detecting smaller larvae's and ants in very early stage of birth period. The main aim of the paper is to propose a framework using machine learning techniques that can be of low cost and efficient system for fish counting and growth study. The expected final result will be a complete preliminary prototype with basic camera setup, focus on larval fish. medium term, improve camera setup and quality; focus on larval and juvenile fish. For the Long term, full fish growth tracking and data mining is implemented. The proposed research in this paper will assist the Malaysian fisheries department to have accuracy on detecting the larvae, juvenile and ants in the hatcheries.

The ability to recognize a grocery on the shelf of a retail store is an ordinary human skill. Automatic detection of grocery on the shelf of retail store provides enhanced value-added to consumer experience, commercial benefits to retailers and efficient monitoring to domestic enforcement ministry. Compared to machine visionbased object recognition system, automatic detection of retail grocery in a store setting has lesser number of successful attempts. In this paper, we present an enhanced YOLOv4 for grocery detection and recognition. We enhanced through spatial pyramid pooling (SPP) and Intersection over union (IOU) components of YOLOv4 to be more accurate in making recognition and faster in the process. We carried an experiment using modified YOLOv4 algorithm to work with our new customized annotated dataset consist on 12000 images with 13 classes. The experiment result shows satisfactory detection compare to other similar works with mAP of 79.39, IoU threshold of 50%, accuracy of 82.83% and real time performance of 61 frames per second

Abstract— Image classification of land use using aerial scene classification has become increasingly common around the world. Utilizing the power of Convolutional Neural Networks (CNNs), identification of various city township areas using satellite imagery has become more efficient compared to the previous manual labeling. The objective of this research is to build a convolutional neural network model for residential and commercial area identification. In the research, we also adopted Inception V3 and VGG16 to develop two transfer learning models for the identification system. The Inception V3-based model achieved the highest overall accuracy value of 100%, showing its effectiveness in accurate residential and commercial area identification. The proposed CNN model achieved an accuracy of 99%, while the VGG-16 model with all configurations being frozen achieved 99% accuracy.

One of the main principle goals for threat protection is to understand the behavior of the employee. An employee who is trusted will have the potential to cause more harm to the organization by collapsing the stability of the computing systems. Thus, insider threat is one of the major security flaws and is very hard to overcome. Currently, in Royal Malaysian Police (RMP) organization, there is a lack of a framework to access such insider threats in their daily operations as well as no data dissemination Standard Operation Policies. Hence, the objective of this research study is to identify the insider threat factors in order to propose best practices, guidelines, policies and security procedures to effectively manage them. This research adapted a mix method approach which is the quantitative methodology in phase one followed by the qualitative method accomplished through semi-structured interviews in a second phase. In the quantitative phase, all the factors were verified and validated by distributing the survey questionnaire to 384 participants from different locations which include the RMP officer, security specialist and administrative officers. The outcome of quantitative data collection was categorized into eight specific influencing factors. These eight factors were further validated through an interview session with 50 participants. The identified factors will be used for design and development of information security best design practices and also as mitigation methods to reduce insider threat risk in the RMP organization.

This paper describes techniques to perform efficient and accurate recognition in larvae images captured from the hatcheries for counting the live and dead larvae’s. In order to accurately model small, irregularly shaped larvae and juvenile, the larvae images are enhanced by three enhancement methods, and segmentation of larvae and juvenile is performed by orientation associated with each edge pixel of region growing segmentation method. The two vital tasks in image analysis are recognition and extraction of larvae and juvenile from an image. When these tasks are manually performed, it calls for human experts, making them more time consuming, more expensive and highly constrained. These negative factors led to the development of various computer systems performing an automatic recognition and extraction of visual information to bring consistency, efficiency and accuracy in image analysis. This main objective of this paper is to study on the various existing automated approaches for recognition and extraction of objects from an image in various scientific and engineering applications. In this study, a categorization is made based on the four principle factors (Input, Segment the larvae, Recognition, Counting) with which each approach is drive .The achieved result of recognition and classification of larvae is around 85%.All the results achieved through matlab implementation are discussed in this paper are proved to work efficiently in real environment.

The most common problem in the daily lives of men and woman is the occurrence of kidney stone, which is named as renal calculi, due to living nature of the people. These calculi can be occurred in kidney, urethra or in the urinary bladder. Most of the existing study in the diagnosis of ultrasound image of the kidney stone identifies the presence or absence of stone in the kidney. The main objective of the paper is to propose a computer aided diagnosis prototype for early detection of kidney stones which helps to change the diet condition and prevention of stone formation in future. The proposed work is based image acquisition, image enhancement, segmentation, feature extraction and classification, whereas in initial stage, ultrasound of kidney image is diagnosed for the presence of renal calculi stone and its level of growth measured in sizes. Seed pixel based region growing segmentation is applied in our work to localize the intensity threshold variation, based on the different threshold variation, it is categorized into a class of images as normal, stone and stone at early stages. The proposed segmentation is based on identifying the homogeneous regions which depends on the granularity features, therefore interested structure with different dimensions are compared with speckle size and extracted. The shape and different size of the grown regions are depending on the entries in lookup table. After completing the stage of region growing, region merging is used to suppress the high frequency artifacts in the ultrasound image. Once the segmented portion of stone is extracted and statistical features are calculated, which can be feed as feature selection by principle component analysis method. The extracted features are in input for artificial neural network classifier for achieving the improved accuracy compared to previous works. The expected output findings are based on texture feature values, threshold variations, size of the stone from the ultrasound kidney image samples with the support of clinical research center. The findings in our study and observation are based on correct estimate size of the stones, position of the stones in location of kidney; these findings are not performed in previous work. The enhanced seed pixel region growing segmentation and ANN classification helps to diagnose the presence or absence of renal calculi kidney stones, which leads to an early detection stone formation in the kidney and improve the accuracy rate of classification.