Mixture of Deterministic and Stochastic Quantization Schemes for Lightweight CNN

We propose Real Full Binary Neural Network (RFBNN), a method that can reduce the memory and compute power of Deep Neural Networks. This method has similar performance to other BNNs in image classification and object detection, while reducing computation power and memory size. In RFBNN, the weight filters are approximated as a binary value by applying a sign function; we apply real binary weight to the whole layer. Therefore, RFBNN can be efficiently implemented on CPU, FPGA and GPU. Results of the all experiments show that the proposed method works successfully on various task such as image classification and object detection. All layers in our networks are composed of only {1, −1}, and unlike the other BNN, there is no scaling factor. Compared to recent network binarization methods, BC, BWN and BWRN, we have reduced the memory size and computation costs, but the performance is the same or better.

Wheat is a primary carbohydrate-rich food commodity in the world, and the health of the wheat plants influences its production. Farmers can face a challenge due to decreased productivity caused by infected plants. However, wheat farmers have used conventional methods to survey diseases; however, these methods are often ineffective and inefficient, taking a lot of time and labor. And for enhanced wheat production, new strategies must be employed. Deep learning, specifically computer vision-based techniques, has proven significant capability in tasks like image classification, segmentation, and object detection. Deep learning techniques such as You Only Look Once (YOLO) models are state-of-the-art neural network algorithms used for accurate object detection. this study presents a comparative evaluation of two state-of-the-art object detection models, YOLOv5 and YOLOv8, for disease detection. Data augmentation techniques such as image noise, rotation, and flipping were implemented to improve the model’s performance during the training phase. The model’s performance was evaluated using metrics such as precision, recall, F1-score, and mean Average Precision (mAP).. The results show that the YOLOv5 and YOLOV8 models achieved good performance. They were able to detect the healthy and disease in images, These findings demonstrate that while both models are highly effective, but YOLOv8 offers greater robustness and accuracy for real-time detection.

Review of deep learning-based methods for non-destructive evaluation of agricultural products

Deep (machine) learning in recent years has significantly increased the predictive modeling strengths for applications in the areas of computer vision, speech recognition, and natural language processing. Being heavily data-driven, conventional general-purpose deep learning methods mainly rely on training data to learn task-specific data representations. In this thesis, we focus on leveraging novel learning algorithms in deep learning to tackle visual recognition challenges at the three key levels -- namely pixel, object, and image levels. Throughout the thesis, we demonstrate how each and every of our proposed methods can address a specific visual recognition problem. Firstly, we apply deep learning to improve the saliency detection performance of deep networks. A coarse saliency map of the input image's salient object is initially predicted by a convolutional-deconvolutional network (or equivalently a fully convolutional network). Following that, the coarse saliency map is sequentially refined by a separate network called Recurrent Attentional Convolutional-Deconvolutional Network (RACDNN) that shares its network weights across its sequential steps. At every sequential (temporal) step, the same RACDNN is applied to a newly selected/attended image sub-region and it performs saliency map refinement for that particular sub-region. Despite having many learnable network parameters and no access to additional training data, RADCNN can be trained well and it outperforms the "single-step" (coarse saliency detection) baseline and state-of-the-art saliency detection methods by large margins, on several saliency detection benchmark datasets. Secondly, we introduce auto-encoder weight transfer networks (AE-WTN+), a deep model for scaling up object detection. Conventional WTN transfers object class knowledge from the classification weights of a pretrained large-scale image classification network, to the classification weights of an object detection network. However, they suffer from under-fitting and are not able to preserve the rich object class information of novel classes. In this work, we propose a new WTN method -- AE-WTN+ which makes use of normalization techniques to ease network optimization. Inputs and intermediate features are normalized to improve trainability for reduced under-fitting. More importantly, a novel auto-encoding (reconstruction) loss is proposed to encourage information preservation of all classes in the weight predictions of AE-WTN+. This enables the predicted weights of AE-WTN+ to maintain the class discriminability and neighbourhood relationships contained in the pre-trained large-scale classification weights. Experiments on large-scale detection datasets validate the effectiveness of AE-WTN+ on both seen and novel classes. Thirdly, we improve the efficiency of cross-layer network connections for very deep convolutional network in the context of image classification. This gives rise to a new convolutional network architecture called Deluge Networks (DelugeNets) which pass the output (feature activations) of any preceding composite network layer as direct inputs to all succeeding composite layers. For the incoming feature activations of preceding composite layers, the succeeding composite layers first apply rather lightweight cross-layer depthwise convolutional weights on them for feature aggregation. The heavy lifting is carried out by the last convolutional weight layer in each preceding composite layer that performs cross-channel convolutions. The proposed technique enables unimpeded information flow among many layers in very deep DelugeNets with great efficiency. Experiments on small-scale and large-scale datasets show that DelugeNets perform competitively to state-of-the-art deep convolutional networks at lower computational and parameter costs. Fourthly, we propose Stochastic Downsampling (SDPoint), a method to overcome a major limitation suffered by conventional deep convolutional networks -- their computational costs during inference are fixed and cannot be changed. SDPoint downsamples the feature maps at a randomly selected point (layer index) in the network hierarchy during training. The many downsampling configurations known as SDPoint instances each entail a unique downsampling point and downsampling ratio. The same convolutional network weights are shared across the different SDPoint instances that downsample feature maps differently. During inference, the computational cost of a network trained with SDPoint can be instantaneously adjusted to fit a given computational budget, by manually selecting an appropriate SDPoint instance. Additionally, sharing weights between different feature map scales provides significant regularization benefits, making the networks trained with SDPoint less sensitive to scales and more competent. In image classification experiments, SDPoint demonstrates significant cost-accuracy and computational advantages, over independently-trained (non-weight-shared) baselines with various inference costs.

Optical character recognition (OCR) is an important branch in the field of machine vision. It involves pattern recognition, image processing, digital signal processing, artificial intelligence and other disciplines. It is a comprehensive. It has important use value and theoretical significance in high-tech fields such as word information processing, office automation, machine translation and real-time monitoring system. In the 21st century, with the popularity of smart phones with high-definition cameras, OCR has a new pursuit in its development: more and more people pick up their mobile phones to photograph the things and scenes they see and obtain the text information in the pictures. Therefore, the recognition of characters in natural scenes has become a brand-new topic. In the past, text detection and text recognition algorithms were basically based on artificially designed features and traditional image processing methods. These features and algorithms were difficult to design and needed a lot of professional knowledge and experience support, so the accuracy was not high and they were not generalized. In recent years, with the rapid development of deep learning technology, breakthroughs have been made in the fields of computer vision such as image classification, object detection and semantic segmentation. Deep learning algorithm is a data-driven algorithm. The algorithm based on deep learning can automatically discover and learn the hidden feature rules in a large number of data through iterative training, without too much human intervention, so it has better generalization than traditional image processing related algorithms.

Traffic violation apprehension is one of the traffic problems here in the Philippines. One example is the No Helmet No Ride Law that is implemented but many motorists still choose to ignore. To alleviate the problem the government has offered many solutions, one of which is the No Contact Traffic Apprehension Policy that uses CCTV Monitoring. To further enhance this solution the government has partnered with the De La Salle University to use artificial intelligence in the system. Computer Vision tasks like image classification and object detection can help automate the traffic apprehension system. Image classification and object detection are technologies which are used in computer vision in defining an image or coordinates of an object in an image. In this work, a novel approach to classifying motorcycle riders between wearing a helmet or not will be developed. It will be demonstrated using deep machine learning, specifically convolutional neural network and by utilizing different pre-trained models to a gathered dataset.

Deep learning techniques have recently demonstrated remarkable success in numerous domains. Typically, the success of these deep learning models is measured in terms of performance metrics such as accuracy and mean average precision (mAP). Generally, a model's high performance is highly valued, but it frequently comes at the expense of substantial energy costs and carbon footprint emissions during the model building step. Massive emission of CO2 has a deleterious impact on life on earth in general and is a serious ethical concern that is largely ignored in deep learning research. In this article, we mainly focus on environmental costs and the means of mitigating carbon footprints in deep learning models, with a particular focus on models created using knowledge distillation (KD). Deep learning models typically contain a large number of parameters, resulting in a 'heavy' model. A heavy model scores high on performance metrics but is incompatible with mobile and edge computing devices. Model compression techniques such as knowledge distillation enable the creation of lightweight, deployable models for these low-resource devices. KD generates lighter models and typically performs with slightly less accuracy than the heavier teacher model (model accuracy by the teacher model on CIFAR 10, CIFAR 100, and TinyImageNet is 95.04%, 76.03%, and 63.39%; model accuracy by KD is 91.78%, 69.7%, and 60.49%). Although the distillation process makes models deployable on low-resource devices, they were found to consume an exorbitant amount of energy and have a substantial carbon footprint (15.8, 17.9, and 13.5 times more carbon compared to the corresponding teacher model). The enormous environmental cost is primarily attributable to the tuning of the hyperparameter, Temperature (τ). In this article, we propose measuring the environmental costs of deep learning work (in terms of GFLOPS in millions, energy consumption in kWh, and CO2 equivalent in grams). In order to create lightweight models with low environmental costs, we propose a straightforward yet effective method for selecting a hyperparameter (τ) using a stochastic approach for each training batch fed into the models. We applied knowledge distillation (including its data-free variant) to problems involving image classification and object detection. To evaluate the robustness of our method, we ran experiments on various datasets (CIFAR 10, CIFAR 100, Tiny ImageNet, and PASCAL VOC) and models (ResNet18, MobileNetV2, Wrn-40-2). Our novel approach reduces the environmental costs by a large margin by eliminating the requirement of expensive hyperparameter tuning without sacrificing performance. Empirical results on the CIFAR 10 dataset show that the stochastic technique achieves an accuracy of 91.67%, whereas tuning achieves an accuracy of 91.78%-however, the stochastic approach reduces the energy consumption and CO2 equivalent each by a factor of 19. Similar results have been obtained with CIFAR 100 and TinyImageNet dataset. This pattern is also observed in object detection classification on the PASCAL VOC dataset, where the tuning technique performs similarly to the stochastic technique, with a difference of 0.03% mAP favoring the stochastic technique while reducing the energy consumptions and CO2 emission each by a factor of 18.5.

Deep neural networks (DNNs) play a crucial role in image classification and object detection, with applications in autonomous driving, security, Unmanned Aerial Vehicle (UAV) navigation, and robotics. Ensuring secure execution on mobile devices is challenging due to the sensitivity of input data and DNN architectures. Hardware-based Trusted Execution Environments (TEEs), such as ARM TrustZone, offer security but face resource and performance limitations when handling full-scale DNNs. This work introduces EffTEE, a security framework that enables efficient DNN execution within mobile TEEs. EffTEE employs dynamic suppression to prune unimportant neurons, foundational neuron restructuring to optimize memory usage, and dynamic slicing for effective model partitioning. Experimental results show that EffTEE reduces inference time by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2 - 60 \times $ </tex-math></inline-formula> while maintaining accuracy comparable to existing secure DNN methods. These findings demonstrate EffTEE’s potential for secure and efficient DNN deployment in resource-constrained environments.

Image classification is one of the important problems in the field of machine learning. Deep learning architectures are used in many machine learning applications such as image classification and object detection. The ability to manipulate large image clusters and implement them quickly makes deep learning a popular method in classifying images. This study points out the success of the convolutional neural networks which is the architecture of deep learning, in solving image classification problems. In the study, the convolutional neural network model of the winner of ilsvrc12 competition is implemented. The method distinguishes 1.2 million images with 1000 categories in success. The application is performed with the caffe library, and the image classification process is employed. In the application that uses the speed facility provided by GPU, the test operation is performed by using the images in Caltech-101 dataset.

Purpose This study aims to automate the visual inspection of piling sheets in water channel construction using artificial intelligence (AI). By employing image classification and object detection techniques, the research focuses on extracting and analysing geometric features to enhance the accuracy and efficiency of the inspection process. It also addresses key challenges associated with the unique characteristics of construction materials and the limited variability of available inspection datasets. Design/methodology/approach Convolutional neural networks (CNNs) with varying complexities are employed for image classification, across four and six classes, and for object detection of piling sheets in water channel environments. A dataset provided by Witteveen + Bos is preprocessed to generate training sets, and the CNN architectures are optimized for enhanced performance. The accuracy and efficiency of the proposed models are evaluated and compared against traditional manual inspection methods. Findings The AI-driven approach significantly reduces processing time, evaluating 40,000 images in just 11.9 h, compared to approximately one month using manual assessment. The 4-class classification model achieves an accuracy of 96%, while the 6-class model attains 72%. The object detection model produces a mean average precision (mAP) of 79%. These results meet the performance standards set by the Dutch company Witteveen + Bos, which demonstrate the effectiveness of AI in automating the inspection of piling sheets. Originality/value This study introduces a novel AI-based approach for assessing piling sheets, demonstrating substantial improvements over traditional inspection methods. It introduces a systematic evaluation of various CNN architectures and hyperparameters to optimize the models specifically for piling sheet inspection rather than relying on off-the-shelf solutions. The use of CNNs for both image classification and object detection adheres to relevant Dutch engineering standards. Notably, the reduction in processing time, from one month to around 12 h, represents a major advancement in the efficiency of civil engineering inspections.

Over the past decade, Deep Learning (DL) techniques have demonstrated remarkable advancements across various domains, driving their widespread adoption. Particularly in medical image analysis, DL received greater attention for tasks like image segmentation, object detection, and classification. This paper provides an overview of DL-based object recognition in medical images, exploring recent methods and emphasizing different imaging techniques and anatomical applications. Utilizing a meticulous quantitative and qualitative analysis following PRISMA guidelines, we examined publications based on citation rates to explore into the utilization of DL-based object detectors across imaging modalities and anatomical domains. Our findings reveal a consistent rise in the utilization of DL-based object detection models, indicating unexploited potential in medical image analysis. Predominantly within Medicine and Computer Science domains, research in this area is most active in the US, China, and Japan. Notably, DL-based object detection methods have gotten significant interest across diverse medical imaging modalities and anatomical domains. These methods have been applied to a range of techniques including CR scans, pathology images, and endoscopic imaging, showcasing their adaptability. Moreover, diverse anatomical applications, particularly in digital pathology and microscopy, have been explored. The analysis underscores the presence of varied datasets, often with significant discrepancies in size, with a notable percentage being labeled as private or internal, and with prospective studies in this field remaining scarce. Our review of existing trends in DL-based object detection in medical images offers insights for future research directions. The continuous evolution of DL algorithms highlighted in the literature underscores the dynamic nature of this field, emphasizing the need for ongoing research and fitted optimization for specific applications.

CMOS Image sensors play a vital role in the exponentially growing field of Artificial Intelligence (AI). Applications like image classification, object detection and tracking are just some of the many problems now solved with the help of AI, and specifically deep learning. In this work, we target image classification to discern between six categories of fruits — fresh/ rotten apples, fresh/ rotten oranges, fresh/ rotten bananas. Using images captured from high speed CMOS sensors along with lightweight CNN architectures, we show the results on various edge platforms. Specifically, we show results using ON Semiconductor’s global-shutter based, 12MP, 90 frame per second image sensor (XGS-12), and ON Semiconductor’s 13 MP AR1335 image sensor feeding into MobileNetV2, implemented on NVIDIA Jetson platforms. In addition to using the data captured with these sensors, we utilize an open-source fruits dataset to increase the number of training images. For image classification, we train our model on approximately 30,000 RGB images from the six categories of fruits. The model achieves an accuracy of 97% on edge platforms using ON Semiconductor’s 13 MP camera with AR1335 sensor. In addition to the image classification model, work is currently in progress to improve the accuracy of object detection using SSD and SSDLite with MobileNetV2 as the feature extractor. In this paper, we show preliminary results on the object detection model for the same six categories of fruits.

Review on self-supervised image recognition using deep neural networks

The building industry is strongly reliant on materials, which account for 55%-60% of its expenses. However, due to outdated and time-consuming approaches, inadequate inventory management prevails. This is where developing technologies such as Deep Learning (DL) might help uncover solutions. Surprisingly, very little scientific research on DL has been conducted for this purpose. As a result, this study looks into the prospect of automating construction warehouse management by employing CNN for object detection and counting. During the initial investigation, 23 studies out of 26 used Convolutional Neural Networks (CNN) for image processing and object detection. Secondly, a model was developed and compared its accuracy to that of human counting and discovered that the model outperformed people. Industry professionals were interviewed to discuss the findings. Industry professionals highlighted the advantages and disadvantages of using such an automated system in construction warehouses. In conclusion, this study shows that the CNN base model outperforms people in counting materials, and the proposed automated inventory management system has significant industry potential.

The objective of this thesis was to evaluate the viability of implementation of an object recognition algorithm driven by deep learning for aerospace manufacturing, maintenance and assembly tasks. Comparison research has found that current computer vision methods such as, spatial mapping was limited to macro-object recognition because of its nodal wireframe analysis. An optical object recognition algorithm was trained to learn complex geometric and chromatic characteristics, therefore allowing for micro-object recognition, such as cables and other critical components. This thesis investigated the use of a convolutional neural network with object recognition algorithms. The viability of two categories of object recognition algorithms were analyzed: image prediction and object detection. Due to a viral epidemic, this thesis was limited in analytical consistency as resources were not readily available. The prediction-class algorithm was analyzed using a custom dataset comprised of 15 552 images of the MaxFlight V2002 Full Motion Simulator’s inverter system, and a model was created by transfer-learning that dataset onto the InceptionV3 convolutional neural network (CNN). The detection-class algorithm was analyzed using a custom dataset comprised of 100 images of two SUVs of different brand and style, and a model was created by transfer-learning that dataset onto the YOLOv3 deep learning architecture. The tests showed that the object recognition algorithms successfully identified the components with good accuracy, 99.97% mAP for prediction-class and 89.54% mAP. For detection-class. The accuracies and data collected with literature review found that object detection algorithms are accuracy, created for live -feed analysis and were suitable for the significant applications of AVI and aircraft assembly. In the future, a larger dataset needs to be complied to increase reliability and a custom convolutional neural network and deep learning algorithm needs to be developed specifically for aerospace assembly, maintenance and manufacturing applications.