Duplicate Frames Research Articles

Inappropriate YouTube content detection and classification by using proposed novel auto-determined k-means clustering and PDBRNN architecture

To ensure a safe and pleasant user experience while watching content on YouTube, it is necessary to identify and classify inappropriate content, especially content that is inappropriate for children. In this work, we have concentrated on establishing an efficient system for detecting inappropriate content on YouTube. Most of the work focuses on manual pre-processing; however, it takes too much time, requires manpower support, and is not ideal for solving real-time problems. To address this challenge, we have proposed an automatic preprocessing scheme for selecting appropriate frames and removing unwanted frames such as noise and duplicate frames. For this purpose, we have utilized the proposed novel auto-determined k-means (PADK-means) algorithm. Our PADK-means algorithm automatically determines the optimal cluster count instead of manual specifications. By doing this, we have solved the manual cluster count specification problem in the traditional k-means clustering algorithm. On the other hand, to improve the system’s performance, we utilized the Proposed Feature Extraction (PFE) method, which includes two pre-trained models DenseNet121 and Inception V3 are utilized to extract local and global features from the frame. Finally, we employ a proposed double-branch recurrent network (PDBRNN) architecture, which includes bi-LSTM and GRU, to classify the video as appropriate or inappropriate. Our proposed automatic preprocessing mechanism, proposed feature extraction method, and proposed double-branch RNN classifier yielded an impressive accuracy of 97.9%.

Enhanced FRER Mechanism in Time-Sensitive Networking for Reliable Edge Computing.

Time-Sensitive Networking (TSN) and edge computing are promising networking technologies for the future of the Industrial Internet. TSN provides a reliable and deterministic low-latency communication service for edge computing. The Frame Replication and Elimination for Reliability (FRER) mechanism is important for improving the network reliability of TSN. It achieves high reliability by transmitting identical frames in parallel on two disjoint paths, while eliminating duplicated frames at the destination node. However, there are two problems with the FRER mechanism. One problem is that it does not consider the path reliability, and the other one is that it is difficult to find two completely disjoint path pairs in some cases. To solve the above problems, this paper proposes a method to find edge-disjoint path pairs considering path reliability for FRER in TSN. The method includes two parts: one is building a reliability model for paths, and the other one is computing a working path and a redundant path with the Edge-Disjoint Path Pairs Selection (EDPPS) algorithm. Theoretical and simulation results show that the proposed method effectively improves path reliability while reducing the delay jitter of frames. Compared with the traditional FRER mechanism, the proposed method reduces delay jitter by 15.6% when the network load is 0.9.

Improving the YOLOV7 Algorithm for Object Detection within Recorded Videos

Object detection algorithms play an important role in detecting people in surveillance videos. In recent years, with the rapid development of deep learning, the performance of object detection has improved by leaps and bounds, and the scheme of object detection by the YOLOV7 algorithm has also been born. Traditional object detection methods often fail to achieve a balance between speed and accuracy. To address these issues, in this research, an improved YOLOv7 algorithm performance is proposed to get the best speed-to-accuracy balance compared to state-of-the-art object detection within recorded videos using an effective compression method. This method calculates the difference between frames of video, and by using the zero difference approach by removing the duplicate frames from the recorded video and choosing only the meaningful frames based on many variables, including frame size, frame details, and the distance of the frames, influence the choice of a meaningful frame, and this will reduce the size of the video by eliminating the frames comparable to those chosen. Additionally, any other datasets or pre-trained weights have not been used; YOLOv7 has been exclusively trained on the MS COCO dataset from scratch. In order to ensure the effectiveness of this approach, numerous detection systems are used in this work. Additionally, positive performance results to reduce the processing time required for object detection have been attained.

Frame Duplication Forgery Detection in Surveillance Video Sequences Using Textural Features

Frame duplication forgery is the most common inter-frame video forgery type to alter the contents of digital video sequences. It can be used for removing or duplicating some events within the same video sequences. Most of the existing frame duplication forgery detection methods fail to detect highly similar frames in the surveillance videos. In this paper, we propose a frame duplication forgery detection method based on textural feature analysis of video frames for digital video sequences. Firstly, we compute the single-level 2-D wavelet decomposition for each frame in the forged video sequences. Secondly, textural features of each frame are extracted using the Gray Level of the Co-Occurrence Matrix (GLCM). Four second-order statistical descriptors, Contrast, Correlation, Energy, and Homogeneity, are computed for the extracted textural features of GLCM. Furthermore, we calculate four statistical features from each frame (standard deviation, entropy, Root-Mean-Square RMS, and variance). Finally, the combination of GLCM’s parameters and the other statistical features are then used to detect and localize the duplicated frames in the video sequences using the correlation between features. Experimental results demonstrate that the proposed approach outperforms other state-of-the-art (SOTA) methods in terms of Precision, Recall, and F1Score rates. Furthermore, the use of statistical features combined with GLCM features improves the performance of frame duplication forgery detection.

Performance Analysis of Overcurrent Protection under Corrupted Sampled Value Frames: A Hardware-in-the-Loop Approach

The IEC 61850 standard aims at digitization substations and provides interoperability between various Intelligent Electronic Device vendors. The digitization process is accompanied by several challenges related to data transmission on the ethernet network and the protection behavior under these conditions. Among the challenges, we can mention packet loss, delay, and duplicate frame, which occurs when the merging units (publisher) transmit the sampled values and, for some reason, these packets do not reach the subscriber or are duplicated. Nowadays, most Intelligent Electronic Device manufacturers block the protection function when some sampled value packets are corrupted. The effects of blocking protection when packet loss occurs under normal operating conditions do not cause significant problems. However, when a fault occurs, the corrupted packets can cause a delay in fault clearance, causing even more damage to the grid. The purpose of this article is to present the effects of corrupted sampled values on the performance of overcurrent protection. All the evaluations were performed in real time using the hardware-in-loop simulation approach with a commercial Intelligent Electronic Device. The OP5700 hardware platform from OPAL-RT, with the library “IEC 61850 Data Integrity Manipulation”, was used. The results show that corrupted sampled value frames affect the functioning of the protections.

Effective and Efficient Video Compression by the Deep Learning Techniques

Deep learning has reached many successes in Video Processing. Video has become a growing important part of our daily digital interactions. The advancement of better resolution content and the large volume offers serious challenges to the goal of receiving, distributing, compressing and revealing high-quality video content. In this paper we propose a novel Effective and Efficient video compression by the Deep Learning framework based on the flask, which creatively combines the Deep Learning Techniques on Convolutional Neural Networks (CNN) and Generative Adversarial Networks (GAN). The video compression method involves the layers are divided into different groups for data processing, using CNN to remove the duplicate frames, repeating the single image instead of the duplicate images by recognizing and detecting minute changes using GAN and recorded with Long Short-Term Memory (LSTM). Instead of the complete image, the small changes generated using GAN are substituted, which helps with frame-level compression. Pixel wise comparison is performed using K-nearest Neighbours (KNN) over the frame, clustered with K-means and Singular Value Decomposition (SVD) is applied for every frame in the video for all three colour channels [Red, Green, Blue] to decrease the dimension of the utility matrix [R, G, B] by extracting its latent factors. Video frames are packed with parameters with the aid of a codec and converted to video format and the results are compared with the original video. Repeated experiments on several videos with different sizes, duration, Frames per second (FPS), and quality results demonstrated a significant resampling rate. On normal, the outcome delivered had around a 10% deviation in quality and over half in size when contrasted, and the original video.

Evaluating Image Redundancies by Analyzing Pixel Based Similarity Ratios

While bifurcating a video at a specific rate and extracting frames from it, there might be an occurrence of repetitive similar frames which can cause redundancy of data. As a result, computation time increases, thus causing problems in determining the most accurate frame for further computation such as content extraction, feature extraction, etc. The intent of this paper is to introduce a solution to detect such lightly identical frames from the input video as well as to detect similar images having equal dimensions from a database. The proposed detection algorithm in this research paper mainly focuses on converting frames to NumPy array objects and then performing array-based numerical operations on them, which also involves comparing a pair of images pixel-by-pixel to determine the similarity ratio between them. It returns a new dataset of frames having minimal duplicate frames, and a compressed version of the original video can be re-constructed using this dataset of frames. The key concept is to eradicate anomalies caused by the redundancy of frames.

Frame Duplication Forgery Detection and Localization Algorithm Based on the Improved Levenshtein Distance

In this digital era of technology and software development tools, low-cost digital cameras and powerful video editing software (such as Adobe Premiere, Microsoft Movie Maker, and Magix Vegas) have become available for any common user. Through these softwares, editing the contents of digital videos became very easy. Frame duplication is a common video forgery attack which can be done by copying and pasting a sequence of frames within the same video in order to hide or replicate some events from the video. Many algorithms have been proposed in the literature to detect such forgeries from the video sequences through analyzing the spatial and temporal correlations. However, most of them are suffering from low efficiency and accuracy rates and high computational complexity. In this paper, we are proposing an efficient and robust frame duplication detection algorithm to detect duplicated frames from the video sequence based on the improved Levenshtein distance. Extensive experiments were performed on some selected video sequences captured by stationary and moving cameras. In the experimental results, the proposed algorithm showed efficacy compared with the state-of-the-art techniques.

Duplicate Frame Video Forgery Detection Using Siamese-based RNN

Video and image data is the most important and widely used format of communication today. It is used as evidence and authenticated proof in different domains such as law enforcement, forensic studies, journalism, and others. With the increase of video applications and data, the problem of forgery in video and images has also originated. Although a lot of work has been done on image forgery, video forensic is still a challenging area. Videos are manipulated in many ways. Frame insertion, deletion, and frame duplication are a few of the major challenges. Moreover, in the perspective of duplicated frames, frame rate variation and loop detection are also key issues. Identification of forged duplication frames for large videos with variant frame rates in real-time is not applicable due to computational limitations, lack of generalization, and low-performance accuracy. This research has investigated the problem of frame duplication with varied frame rates using a deep learning approach. A novel deep learning framework consisting of Inflated 3D (I3D) and Siamese-based Recurrent Neural Network (RNN) is proposed to resolve the aforementioned issues. The first step in the proposed framework is to extract the features and convert videos into frames. I3D network receives an original and a forged video to detect frame-to-frame duplication. Then multiple frames are merged to create a sequence. This sequence is passed to Siamese-based RNN which is used for the sequence to sequence forgery detection in video. Media Forensic Challenge (MFC) is a relatively new dataset with various frame rates, and a huge volume of videos. MFC and Image Retrieval and Analysis Tool (VIRAT) datasets are used for training and validation of the proposed model. The accuracy of the proposed method with the VIRAT dataset is 86.6% and with the MFC dataset 93%. The comparative analysis with state-of-the-art approaches has shown the robustness of the proposed approach.

Real-Time Violent Action Recognition Using Key Frames Extraction and Deep Learning

Violence recognition is crucial because of its applications in activities related to security and law enforcement. Existing semi-automated systems have issues such as tedious manual surveillances, which causes human errors and makes these systems less effective. Several approaches have been proposed using trajectory-based, non-object-centric, and deep-learning-based methods. Previous studies have shown that deep learning techniques attain higher accuracy and lower error rates than those of other methods. However, the their performance must be improved. This study explores the state-of-the-art deep learning architecture of convolutional neural networks (CNNs) and inception V4 to detect and recognize violence using video data. In the proposed framework, the keyframe extraction technique eliminates duplicate consecutive frames. This keyframing phase reduces the training data size and hence decreases the computational cost by avoiding duplicate frames. For feature selection and classification tasks, the applied sequential CNN uses one kernel size, whereas the inception v4 CNN uses multiple kernels for different layers of the architecture. For empirical analysis, four widely used standard datasets are used with diverse activities. The results confirm that the proposed approach attains 98% accuracy, reduces the computational cost, and outperforms the existing techniques of violence detection and recognition.

A passive blind forgery detection technique to identify frame duplication attack

This paper presents passive blind forgery detection to identify frame duplication attack in Moving Picture Experts Group-4 (MPEG-4) video. In this attack, one or more frames are copied and pasted at other location in the same video to hide or highlight particular activity. Since the tampered frames are from the same video, their statistical properties are uniform, which makes challenging to identify duplicate frames. In this paper, a two-step algorithm is proposed, in which the suspicious frames are identified, their features are extracted and compared with other frames of the test video to take the decision. Scale Invariant Feature Transform (SIFT) key-points are used as feature for comparison. Finally, Random Sample Consensus algorithm is used to locate duplicate frames. The proposed method is tested on compressed and uncompressed videos with variable compression rate. The simulation results show that the proposed scheme is competent to detect the tampered frames with 99.8% average accuracy. Comparative analysis is made for the proposed method with existing methods with respect to parameters Precision Rate (PR), Recall Rate (RR), Detection Accuracy (DA). The average values of PR, RR, and DA for the proposed method are 99.9%, 99.7%, 99.8% respectively, which are better than other methods. The proposed method needs average 33 seconds of simulation time, which is less as compared to other methods.

Video Forgery Detection Method based on Local Difference Binary

Recently, the rapid development of video editing software has made video forgery applicable. Researchers have proposed methods to detect forged video frames. These methods utilize codec properties, motion artifacts, noise effect and frame similarity to detect forgery. Execution time and low detection accuracy are the two main drawbacks of forgery detection methods reported in the literature. In this study, a new frame duplication detection method using Local Difference Binary (LDB) is proposed to extract features from the frames. Distance between similar frames that have similar feature vectors are is used by the method to estimate Distance of Forgery and to determine the exact location of duplicated frames. PSNR between similar frames are is then used to group them into three classes, and rule-based mechanism reports forged frames according to the membership to classes. Experimental results indicate that the proposed method has lower execution time with higher accuracy than similar works.

An Efficient Relayed Broadcasting Based on the Duplication Estimation Model for IoT Applications

In this paper, we consider relay-based broadcasting in wireless ad hoc networks, which can enable various emerging services in the Internet of Things (IoT). In this kind of traffic dissemination scheme, also known as flooding, all the nodes not only receive frames but also rebroadcast them. However, without an appropriate relay suppression, a broadcast storm problem arises, i.e., the transmission may fail due to severe collisions and/or interference, many duplicate frames are unnecessarily transmitted, and the traffic dissemination time increases. To mitigate the broadcast storm problem, we propose a reasonable criterion to restrict the rebroadcasting named the duplication ratio. Based on this, we propose an efficient mechanism consisting of duplication suppression and re-queuing schemes. The former discards duplicate frames proactively in a probabilistic manner to decrease the redundancy whereas the latter provides a secondary transmission opportunity reactively to compensate for the delivery failure. Moreover, to apply the duplication ratio practically, we propose a simple method to approximate it based on the number of adjacent nodes. The simulation study confirms that the proposed mechanism tightly ensured the reliability and decreased the traffic dissemination time by up to 6-fold compared to conventional mechanisms.

Patient necessity notification system based on gesture recognition (Kinect V2) and internet of things using selection frame method

The use of gesture recognition integrated with Internet of Things (IoT) as a system for detecting patient needs provides information speed and accuracy for nurses. One tool that is widely used to implement gesture recognition is Kinect (Version 2). For data communication, Kinect will be collaborated with IoT in the process of sending data from the start after gesture recognition to the presentation of information to end users. The problem that arises when Kinect is integrated with IoT is the burden on the network because the average number of frames read by Kinect in one second is 30 frames. In this study, a frame selection method was developed to minimize the number of frames sent to IoT networks by minimizing the number of duplicate frames. The selection frame method on gesture recognition integrated with IoT is then tested for performance with distance and delay parameters. The system implementation results are able to read 100% gesture recognition, and the accuracy of the information sent by the nurse is 100% in the testing environment. The optimal distance that can provide a reading of more than 95% is 2.5 meters. The optimal value of 100% is not achieved because some testing of the falling position cannot read the whole skeleton.

Visual recognition and classification of videos using deep convolutional neural networks

Classification of videos based on its content is one of the challenging and significant research problems. In this paper, a simple and efficient model is proposed for classification of sports videos using deep learned convolution neural networks. In the proposed research, the gray scale variants of image frames are employed for classification process through convolution technique at varied levels of abstraction by adapting it through a sequence of hidden layers. The image frames considered for classification are obtained after the duplicate frame elimination and each frame is further rescaled to dimension 120x240. The sports videos categories used for experimentation include badminton, football, cricket and tennis which are downloaded from various sources of google and YouTube. The classification in the proposed method is performed with Deep Convolution Neural Networks (DCNN) with around 20 filters each of size 5x5 with around stride length of2 and its outcomes are compared with Local Binary Patterns (LBP), Bag of Words Features (BWF) technique. The SURF features are extracted from the BWF technique and further 80% of strongest feature points are employed for clustering the image frames using K-Means clustering technique with an average accuracy achieved of about 87% in classification. The LBF technique had produced an average accuracy of 73% in differentiating one image frame to other whereas the DCNN had shown a promising outcome with accuracy of about 91% in case of 40% training and 60% test datasets, 99% accuracy in case of 60% training an 40% test datasets. The results depict that the proposed method outperforms the image processing-based techniques LBP and BWF.

A Novel Optical High-Availability Seamless Redundancy (OHSR) Design Based on Beam Splitting / Combining Techniques

The standard high-availability seamless redundancy (HSR) protocol utilizes duplicated frame copies of each sent frame for zero fail over time. This means that even in cases of a node or link failure, the destination node will receive at least one copy of the sent frame, resulting in no network downtime. However, the standard HSR is mostly based on the electrical signal connection inside the node, which leads to the production of considerable latency at each node due to frame processing. Therefore, in a large scale HSR ring network, the accumulated latencies become significant and can often restrict the mission-critical real-time application of HSR. In this paper, we present a novel design for optical HSR (OHSR) that uses beam splitting/combining techniques. The proposed OHSR passes the frames directly to adjacent nodes without frame processing at each node, thereby theoretically generating no latency in any node. Various simulations for network samples, made to validate the OHSR design and its performance, show that the OHSR outperforms the standard HSR.

A Comparison of Techniques for Reducing Unicast Traffic in HSR Networks

This paper investigates several existing techniques for reducing high-availability seamless redundancy (HSR) unicast traffic in HSR networks for substation automation systems (SAS). HSR is a redundancy protocol for Ethernet networks that provides duplicate frames for separate physical paths with zero recovery time. This feature of HSR makes it very suited for real-time and mission-critical applications such as SAS systems. HSR is one of the redundancy protocols selected for SAS systems. However, the standard HSR protocol generates too much unnecessary redundant unicast traffic in connected-ring networks. This drawback degrades network performance and may cause congestion and delay. Several techniques have been proposed to reduce the redundant unicast traffic, resulting in the improvement of network performance in HSR networks. These HSR traffic reduction techniques are broadly classified into two categories based on their traffic reduction manner, including traffic filtering-based techniques and predefined path-based techniques. In this paper, we provide an overview and comparison of these HSR traffic reduction techniques found in the literature. The concepts, operational principles, network performance, advantages, and disadvantages of these techniques are investigated, summarized. We also provide a comparison of the traffic performance of these HSR traffic reduction techniques.

Hadamard based Video Key Frame Extraction using Thepade's Transform Error Vector Rotation with Assorted Similarity Measures

Video summarization is a method to reduce redundancy and generate succinct representation of the video data. In video summarization process, several frames containing similar information need to get processed, this leads to slower processing speed and higher complexity, consuming. More time Video summarization using key frames can ease the speed up of video processing. One of the mechanisms to generate video summaries is to extract key frames which represent the most important content of the video by identifying neard duplicate frames in video. In this paper, novel key frames extraction method is proposed with Thepade's Walsh Hademard Error Vector Rotation (THdEVR) with ten different codebook sizes and and assorted similarity measures. Experimentation done with help of the test bed of videos has shown that higher codebook sizes give better completeness in key frame extraction for video summarization. Experimental results are discussed for video content summarization with five assorted similarity measures like Euclidean Distance, Canberra Distance, Square-Chord Distance, Mean Square Error, Sorensen Distance with proposed THadVR. Keywordsframe, video summarization, vector quantization, hademard

FHT: A Novel Approach for Filtering High-Availability Seamless Redundancy (HSR) Traffic

High-availability seamless redundancy (HSR) is a protocol for Ethernet networks that provides duplicated frames with zero recovery time in the event of any network component’s failure. It is suited for applications that demand high availability and a very short time-outs such as substation automation systems (SAS). However, HSR generates excessive unnecessary unicast frames and spreads them throughout connected-ring networks, whether or not the destination node exists in network’s rings. This unnecessary redundant traffic causes high bandwidth consumption, resulting in degradation of network performance. In this paper, we introduce a novel approach for filtering and reducing HSR unicast traffic in connected-ring networks, called “filtering HSR traffic” (FHT). The purpose of FHT is to filter HSR unicast traffic and remove circulated traffic for all rings in connected-ring networks. Therefore, FHT significantly reduces network unicast traffic in connected-ring networks. The traffic performance of FHT has been analyzed, evaluated, and compared to that of standard HSR protocol and the port locking (PL) approach. Various simulations were conducted to validate the traffic performance analysis. Analytical and simulation results showed that, for our sample network, FHT reduced network unicast traffic by about 82% compared with standard HSR and by about 56% compared with the PL approach, thus freeing up network bandwidth and improving network traffic performance.

PSFCS: Robust Emergency Communications Supporting High Mobility Based on WiMAX MMR Networks

Nowadays public safety networks are widely deployed to support highly reliable wireless communications in case of emergency. The adoption of standardized systems such as WiMAX for emergency communications represents a significant advance in the off-the-shelf technologies for error detection and error correction. Since it is difficult to fully eliminate the Doppler effect under high speed moving environment, we propose an enhanced CRC-based error correction scheme that carries as much extra segmented frame Check Sequence (FCS) information as subblocks of emergency multicast/broadcast service (MBS) frame, called progressive and selective frame check sequence (PSFCS). In a mobile multihop relay (MMR) environment, a high-mobility mobile station might receive an emergency MBS frame more than once, so the PSFCS could be cross-referenced by duplicated frames received. Therefore the PSFCS achieves a significant performance improvement over the legacy FCS scheme in terms of error detection and correction of emergency MBS frames in a WiMAX MMR network operating in transparent mode. The experimental results show that the goodput and utilization of high-mobility mobile stations can be improved remarkably.