No abstract

Smart city is a city which is designed to meet the people's demands. In addition to use of sources efficiently, trends of people are also a need that smart city should meet. Buying personalized products in a cheap and fast way is a demand of people of today. Mass customization, which is defined as the personalization of products, achieves making the tailor-made products cheaper. In this study, we propose a new approach for mass customization with the integration of smart retail and smart production. With removing the operators and actualizing the progress autonomously, it is aimed to reduce the waiting time of customers. Because less waiting time means that there are more mass-customization customers, and this is expected to increase the popularity of mass customization. Thus, reducing wastes and increasing productivity are aimed. This study also constitutes the infrastructure that enables a production system to autonomously perform all stages from order to delivery. With the given scenarios, challenges and advantages of desired approach are discussed.

The resource need for deep learning and quantum computers' high computing power potential encourage collaboration between the two fields. Today, variational quantum circuits are used to perform the convolution operation with quantum computing. However, the results produced by variational circuits do not show a direct resemblance to the classical convolution operation. Because classical data is encoded into quantum data with their exact values in value-encoded methods, in contrast to variational quantum circuits, arithmetical operations can be applied with high accuracy. In this study, value-encoded quantum circuits for convolution and pooling operations are proposed to apply deep learning in quantum computers in a traditional and proven way. To construct the convolution and pooling operations, some modules such as addition, multiplication, division, and comparison are created. In addition, a window-based framework for quantum image processing applications is proposed. The generated convolution and pooling circuits are simulated on the IBM QISKIT simulator in parallel thanks to the proposed framework. The obtained results are verified by the expected results. Due to the limitations of quantum simulators and computers in the NISQ era, the used grayscale images are resized to 8x8 and the resolution of the images is reduced to 3 qubits. With developing the quantum technologies, the proposed approach can be applied for bigger and higher resolution images. Although the proposed method causes more qubit usage and circuit depth compared to variational convolutional circuits, the results they produce are exactly the same as the classical convolution process.

Continuity of production is a highly important in the days that manufacturing is becoming bigger and serial. The mistakes done while producing process cause fail on products and it may bring about even big losses for the facility. Furthermore, hitches on robots at production line may also cause crucial damages that may give rise to high repair costs and discontinuance of production. In this study, it is aimed to obtain alive bird's eye view map of production lines, which are big and impossible to be monitored with only one camera, by using multi cameras and stitching algorithms. Finding the similar scenes of input images, estimation of homography, warping and blending operations, which are the steps used in feature based image-stitching algorithms, are applied respectively on images that are taken by cameras. The assignment of second nearest neighbor distance rate adaptively makes the results more qualified. After obtaining single stitched image movement detection is actualized by using the difference of sequential frames, and anomaly movements are determined. As a result, the robots at the long production lines can be monitored in one screen, and with processing the obtained image, faults on robots that may cause damage at non-cheap machines can be handled before time.

Nowadays, using image processing techniques on the image taken by camera at the top of production line for detection of faulty products is widespread. The quality of images can decline due to the distance between camera and production line while trying to get whole image of the bigger products. Therefore, in this study we employed camera array instead of only one camera and placed it at a nearer position to production line. As the images taken from cameras in camera array contain specific parts of the whole product, it is required to stitch images and obtain single image of the product before analyze it. Production line speed or the product position can be vary so it is required to find the optimal stitching points in order to get single image. As an optimization technique micro genetic algorithm (µGA) which refers to Genetic Algorithm (GA) with small population size and re-initialization process is popular for its proper and fast solutions. In this paper, we have developed a method using µGA known by its high convergence rate and low termination possibility at a local optimum to be used in image stitching process. Using µGA known by its high convergence rate, less need of computational source and rare premature solution provided us optimal and fast solutions. Experimental results show that µGA, which outperforms conventional GA, gives good results in a reasonable time and it can be used in image stitching process as an alternative.

Özet-Görüntü iĢleme, dijital ortamdaki görüntüler üzerinde bir amaç doğrultusunda gerekli analizler veya değiĢikliklerin yapılmasına yönelik çalıĢmaları kapsamaktadır. Son zamanlarda görüntü iĢleme tekniklerinden en çok kullanılanlardan biri de ortak sahneleri içeren resimler üzerinde iĢlem yapan ve bunları çakıĢtırarak birleĢtiren görüntü mozaikleme iĢlemidir. Bu çalıĢmada kendi kameramızdan elde ettiğimiz görüntüler üzerinde MATLAB ortamında görüntü mozaikleme iĢlemi gerçekleĢtirilmiĢtir. Görüntü mozaikleme için, özellik tabanlı görüntü mozaiklemenin temel adımları olan anahtar noktaların tespiti, bu noktaların eĢleĢtirilmesi, hatalı eĢleĢmelerin elenmesi ve homografi tahmini uygulanmıĢtır. Uygulamada özellik çıkarımı için SIFT algoritması seçilmiĢ, elde edilen homografi matrisi yardımı ile gerekli dönüĢüm iĢlemleri uygulanarak görüntü mozaikleme iĢlemi gerçekleĢtirilmiĢtir. Mozaikleme iĢleminin kalitesini artırmak adına optimizasyon yöntemi olan genetik algoritmadan yararlanılmıĢtır. Uygulamada giriĢ görüntü sayıları ve mozaikleme parametreleri farklı seçilerek çeĢitli karĢılaĢtırma iĢlemleri yapılmıĢtır.Anahtar Kelimeler-Genetik algoritma, Görüntü iĢleme, Görüntü mozaikleme, Optimizasyon. Optimization Based Adaptive Image Mosaicing AlgorithmAbstract-Image processing involves studies to make necessary analyzes or changes in an image aimed at digital images. One of the most recent use of image processing techniques is image mosaicking, which processes and combines images containing common scenes. In this study, we performed image mosaicing in the MATLAB environment on the images we acquired from our own camera. The key steps for feature-based image mosaicking which are detection of key points, matching of these points, elimination of mismatches and homography estimation are applied for image mosaicing. In the application, the SIFT algorithm was selected for feature extraction and image transformation was performed with the help of the obtained homography matrix. In order to increase the quality of the mosaic process, genetic algorithm, which is an optimization method, has been used. Various comparisons were performed by taking different input counts and by choosing different image mosaicing parameters.

No abstract

No abstract