images are a vital part of our everyday life and imageprocessing is the heart of all the modern technologies, including machinevision, artificial intelligence, robotics, deep learning. It would not be wrong to say t...
images are a vital part of our everyday life and imageprocessing is the heart of all the modern technologies, including machinevision, artificial intelligence, robotics, deep learning. It would not be wrong to say that imageprocessing is one of the many reasons for achieving success in any industrial domain, whether it be medical, food, textile, or any other automation industry. It is next to impossible to work in these domains without having sufficient knowledge and skills about imageprocessing techniques. In this thesis document you will find the significance of imageprocessing used in three diverse projects. Each one of the projects is described as a separate chapter in this document. The first project is focused on reducing the power consumption in OLED-based devices. Actually there are two main goals of this project, first one, as the name suggests, is to minimize the power consumed by an OLED device to display images, and the second goal is to simultaneously enhance the color contrasts in images. OLED display panels have become increasingly popular in recent years, thanks to their numerous advantages over the traditional LCD displays. Power consumption in OLED displays depends on the contents where as the backlight is responsible for power consumption in LCD displays. This image-dependent or content-dependent power consumption model of OLED displays have encouraged numerous researchers to create possibilities for reducing the power consumption in OLED-based devices. One such possibility has been explored in this Ph. D. research work. Another industrial application has been presented in the second part of the thesis document. It is a part of the "Food Digital Monitoring" project, funded by Regione Piemonte. The major aim of this project is to identify the healthy and contaminated hazelnuts by using fluorescence and spectral imaging techniques. Two types of contamination are discussed in this work, one, caused by bacterial and fungal infections, called "rot
Lateral flow assays (LFAs) are important diagnostic tools with numerous applications in various scientific fields, including diagnostics, medicine, analytical chemistry, biochemistry, environmental and food science. A...
详细信息
Lateral flow assays (LFAs) are important diagnostic tools with numerous applications in various scientific fields, including diagnostics, medicine, analytical chemistry, biochemistry, environmental and food science. Artificial Intelligence (AI) and imageprocessing tools are the state-of-the-art technology in analytical tools, especially in Point-of-Care (POC) devices that improve the detection efficiency without the need for highly qualified personnel. In this context, we have developed novel multicolor LFAs exploiting machinevision and image analysis tools for the automated "reading" of the visual result of LFAs using beads of different colors as reporters to distinguish between multiple targets. The system consists of a multicolor test integrated with a mobile/ smartphone and a web application for the automatic interpretation of the results. The use of multicolor beads, relating each color to a specific target, enhanced image analysis-based discrimination of the tests between different targets. The developed diagnostic tool has been applied to cutting-edge liquid biopsy applications which include the detection of three different microRNA molecules spiked in urine samples. The developed integrated system has been successfully applied to a series of real samples, advancing the field of LFAs diagnostics. The system showed 99.3 % accuracy, 99.1 % sensitivity and 100 % specificity.
This book focuses on the latest developments in the fields of visual AI, imageprocessing and computer vision. It shows research in basic techniques like image pre-processing, feature extraction, and enhancement, alon...
详细信息
ISBN:
(数字)9783110756722;9783110756821
ISBN:
(纸本)9783110756678
This book focuses on the latest developments in the fields of visual AI, imageprocessing and computer vision. It shows research in basic techniques like image pre-processing, feature extraction, and enhancement, along with applications in biometrics, healthcare, neuroscience and forensics. The book highlights algorithms, processes, novel architectures and results underlying machine intelligence with detailed execution flow of models.
machine learning (ML) models have experienced remarkable growth in their application for multimodal data analysis over the past decade [1]. The diverse applications of ML models span domains such as medical image [2,3...
machine learning (ML) models have experienced remarkable growth in their application for multimodal data analysis over the past decade [1]. The diverse applications of ML models span domains such as medical image [2,3,4] and signal processing [5,6], remote sensing for earth observation and monitoring [7,8,9], the detection of daily human activities [10,11], and many more. ML models play a significant role in supporting computer vision and image-sensing applications, helping to unravel complex and real-world challenges. Recent developments in ML empower us to better analyse image and sensor data, motivating extensive research initiatives aimed at addressing applied challenges in multiple domains, including healthcare, agriculture, defence, remote sensing, earth observation, and autonomous navigation.
This paper presents a comprehensive examination of innovative strategies aimed at enhancing machinevision technology, particularly in the context of energy efficiency and processing speed, critical factors for applic...
详细信息
ISBN:
(纸本)9798350376258
This paper presents a comprehensive examination of innovative strategies aimed at enhancing machinevision technology, particularly in the context of energy efficiency and processing speed, critical factors for applications like facial recognition. The study focuses on three distinct approaches: an optimized two-dimensional convolution algorithm, a novel Field-Programmable Gate Array (FPGA) implementation, and advancements in multichannel meta-imagers. Firstly, the paper discusses an optimized algorithm for two-dimensional convolutions, a fundamental operation in machinevision. This advanced algorithm significantly reduces computational complexity. For instance, in executing a two-dimensional 3×3 cyclic convolution, the proposed method reduces the number of necessary multiplications from 81 to merely 13, offering a substantial improvement in efficiency. Secondly, the paper explores an innovative FPGA implementation of the two-dimensional convolution algorithm. This implementation is designed to minimize the use of shift registers, multipliers, and adders. As a result, it utilizes fewer Look-Up Tables (LUTs), leading to energy and time savings in executing the convolution process. The paper details the architecture of this FPGA-based approach and its implications for energy consumption and processing speed in machinevisionapplications. Finally, the paper introduces a novel technique called the Avg-Topk method, addressing a critical challenge in the pooling layer of convolutional neural networks. This method combines the benefits of average pooling with the advantages of max pooling, aiming to enhance the accuracy of the pooling layer without compromising on efficiency. The Avg-Topk method represents a significant step forward in optimizing the pooling process within machinevision systems. In summary, this paper delves into groundbreaking methods to improve the speed and energy efficiency of machinevision systems, offering valuable insights and potential solution
In low-light environments, machinevision tasks often suffer from performance degradation because traditional image Signal processing (ISP) pipelines are primarily optimized for image quality metrics such as Peak Sign...
详细信息
In low-light environments, machinevision tasks often suffer from performance degradation because traditional image Signal processing (ISP) pipelines are primarily optimized for image quality metrics such as Peak Signal-to-Noise Ratio and Structural Similarity Index, which do not adequately address the specific needs of these applications. Existing methods fall short in enhancing the critical image features required for computer vision tasks under challenging lighting conditions. To address this, we introduce PhyDiISP, a physics-guided, differentiable ISP pipeline designed to improve machinevision performance in low-light scenarios. PhyDiISP integrates traditional ISP design principles with physical insights, including demosaicing for RAW-to-RGB conversion, global tone mapping to adjust overall brightness, and Multiscale Retinex-based enhancement to tackle low-light challenges. Experimental results show that PhyDiISP outperforms existing ISP methods in object detection accuracy across standard benchmarks by effectively enhancing key image features. Furthermore, when trained with L1 loss and aligned with ground truth on datasets of dark-light environments and real RAW-to-RGB conversions, it demonstrates competitive image quality. These results confirm that PhyDiISP is a viable and effective solution for real-world low-light machinevisionapplications.
The emergent ecosystems of intelligent edge devices in diverse Internet-of-Things (IoT) applications, from automatic surveillance to precision agriculture, increasingly rely on recording and processing a variety of im...
详细信息
The emergent ecosystems of intelligent edge devices in diverse Internet-of-Things (IoT) applications, from automatic surveillance to precision agriculture, increasingly rely on recording and processing a variety of image data. Due to resource constraints, e.g., energy and communication bandwidth requirements, these applications require compressing the recorded images before transmission. For these applications, image compression commonly requires: 1) maintaining features for coarse-grain pattern recognition instead of the high-level details for human perception due to machine-to-machine communications;2) high compression ratio that leads to improved energy and transmission efficiency;and 3) large dynamic range of compression and an easy tradeoff between compression factor and quality of reconstruction to accommodate a wide diversity of IoT applications as well as their time-varying energy/performance needs. To address these requirements, we propose, MAGIC, a novel machine learning (ML)-guided image compression framework that judiciously sacrifices the visual quality to achieve much higher compression when compared to traditional techniques, while maintaining accuracy for coarse-grained vision tasks. The central idea is to capture application-specific domain knowledge and efficiently utilize it in achieving high compression. We demonstrate that the MAGIC framework is configurable across a wide range of compression/quality and is capable of compressing beyond the standard quality factor limits of both JPEG 2000 and WebP. We perform experiments on representative IoT applications using two vision data sets and show 42.65x compression at similar accuracy with respect to the source. We highlight low variance in compression rate across images using our technique as compared to JPEG 2000 and WebP.
The emergence of the Internet-of-Things is anticipated to create a vast market for what are known as smart edge devices,opening numerous opportunities across countless domains,including personalized healthcare and adv...
详细信息
The emergence of the Internet-of-Things is anticipated to create a vast market for what are known as smart edge devices,opening numerous opportunities across countless domains,including personalized healthcare and advanced *** 3D integration,edge devices can achieve unprecedented miniaturization while simultaneously boosting processing power and minimizing energy ***,we demonstrate a back-end-of-line compatible optoelectronic synapse with a transfer learning method on health care applications,including electroencephalogram(EEG)-based seizure prediction,electromyography(EMG)-based gesture recognition,and electrocardiogram(ECG)-based arrhythmia *** experiments on three biomedical datasets,we observe the classification accuracy improvement for the pretrained model with 2.93%on EEG,4.90%on ECG,and 7.92%on EMG,*** optical programming property of the device enables an ultralow power(2.8×10^(-13) J)fine-tuning process and offers solutions for patient-specific issues in edge computing ***,the device exhibits impressive light-sensitive characteristics that enable a range of light-triggered synaptic functions,making it promising for neuromorphic vision *** display the benefits of these intricate synaptic properties,a 5×5 optoelectronic synapse array is developed,effectively simulating human visual perception and memory *** proposed flexible optoelectronic synapse holds immense potential for advancing the fields of neuromorphic physiological signal processing and artificial visual systems in wearable applications.
Background In many medically developed applications, the process of early diagnosis in cases of pulmonary disease does not exist. Many people experience immediate suffering due to the lack of early diagnosis, even aft...
详细信息
Background In many medically developed applications, the process of early diagnosis in cases of pulmonary disease does not exist. Many people experience immediate suffering due to the lack of early diagnosis, even after becoming aware of breathing difficulties in daily life. Because of this, identifying such hazardous diseases is crucial, and the suggested solution combines computer vision and communication processing techniques. As computing technology advances, a more sophisticated mechanism is required for *** The major objective of the proposed method is to use imageprocessing to demonstrate computer vision-based experimentation for identifying lung illness. In order to characterize all the uncertainties that are present in nodule segments, an improved support vector machine is also integrated into the decision-making *** As a result, the suggested method incorporates an Improved Support Vector machine (ISVM) with a clear correlation between various margins. Additionally, an imageprocessing technique is introduced where all impacted sites are marked at high intensity to detect the presence of pulmonary syndrome. Contrary to other methods, the suggested method divides the imageprocessing methodology into groups, making the loop generation process much *** Five situations are taken into account to demonstrate the effectiveness of the suggested technique, and test results are compared with those from existing *** The proposed technique with ISVM produces 83 percent of successful results.
暂无评论