This paper proposes a novel, lightweight method to generate animated graphical interchange format images (GIFs) using the computational resources of a client device. The method analyzes an acoustic ...feature from the climax section of an audio file to estimate the timestamp corresponding to the maximum pitch. Further, it processes a small video segment to generate the GIF instead of processing the entire video. This makes the proposed method computationally efficient, unlike baseline approaches that use entire videos to create GIFs. The proposed method retrieves and uses the audio file and video segment so that communication and storage efficiencies are improved in the GIF generation process. Experiments on a set of 16 videos show that the proposed approach is 3.76 times more computationally efficient than a baseline method on an Nvidia Jetson TX2. Additionally, in a qualitative evaluation, the GIFs generated using the proposed method received higher overall ratings compared to those generated by the baseline method. To the best of our knowledge, this is the first technique that uses an acoustic feature in the GIF generation process.
Humans can be identified through the way they walk, according to medical studies, each person has an almost unique way of walking. Over the past decade, numerous gait recognition approach has been ...presented by researchers through which human can be identified, but gait recognition can be affected by a variety of wearable objects such as clothes, carry bags, shoes and others. The silhouette-image with invariant features can be obtained if the wearer objects are removed from it. The Generator Advisory Network (GAN) model is used to remove covariate objects from the silhouette images. The main challenge of this approach to make input-output pairs of frames to train GAN model, for this silhouette image has to mapped with another silhouette image with same pose. Since pose analysis has been done in Pose Energy Image (PEI) approach so PEI is used to label the frames with a specific key pose. After labelling silhouette frames, pair them with the recognition of same pose and train the GAN model. The trained GAN model used to generate a new gait invariant feature which is named as Covariate Factors Omitted Silhouette Image (CFOSI). For cross-validation of the proposed approach, three publically available datasets are used, where proposed approach found 5.50, 4.71 and 17.88 percentage more accurate results with respect to CASIA B, CASIA C and TUM-GAID in case of Gait Energy Image (GEI). We can conclude that after the omission of covariate objects from silhouette images our approach performance is more accurate than others.
Colorectal cancer refers to cancer of the colon or rectum; and has high incidence rates worldwide. Colorectal cancer most often occurs in the form of adenocarcinoma, which is known to arise from ...adenoma, a precancerous lesion. In general, colorectal tissue collected through a colonoscopy is prepared on glass slides and diagnosed by a pathologist through a microscopic examination. In the pathological diagnosis, an adenoma is relatively easy to diagnose because the proliferation of epithelial cells is simple and exhibits distinct changes compared to normal tissue. Conversely, in the case of adenocarcinoma, the degree of fusion and proliferation of epithelial cells is complex and shows continuity. Thus, it takes a considerable amount of time to diagnose adenocarcinoma and classify the degree of differentiation, and discordant diagnoses may arise between the examining pathologists. To address these difficulties, this study performed pathological examinations of colorectal tissues based on deep learning. The approach was tested experimentally with images obtained via colonoscopic biopsy from Gyeongsang National University Changwon Hospital from March 1, 2016, to April 30, 2019. Accordingly, this study demonstrates that deep learning can perform a detailed classification of colorectal tissues, including colorectal cancer. To the best of our knowledge, there is no previous study which has conducted a similarly detailed feasibility analysis of a deep learning-based colorectal cancer classification solution.
The proliferation of information technology has prompted researchers to create a multitude of new security solutions for secure electronic applications, especially on the Internet. Among them, ...security officials prefer authentication systems for user’s identity identification. Indeed, biometric authentication has proved to be superior in many respects compared to the traditional authentication means. Unfortunately, these systems are vulnerable to a variety of attacks, the most serious of which is perhaps the attack on the stored or transmitted template, which makes the safety of this template more important in the design of the biometric systems. This research, therefore, suggests an effective feature extraction method that can provide a deep and cancelable biometric feature. In this study, DCTNet deep learning is combined with chaotic systems to extract revocable palmprint/palm-vein features.
Medical images have a very significant impact in the diagnosing and treating process of patient ailments and radiology applications. For many reasons, processing medical images can greatly improve ...the quality of radiologists’ job. While 2D models have been in use for medical applications for decades, wide-spread utilization of 3D models appeared only in recent years. The proposed work in this paper aims to segment medical volumes under various conditions and in different axel representations. In this paper, we propose an algorithm for segmenting Medical Volumes based on Multiresolution Analysis. Different 3D volume reconstructed versions have been considered to come up with a robust and accurate segmentation results. The proposed algorithm is validated using real medical and Phantom Data. Processing time, segmentation accuracy of predefined data sets and radiologist’s opinions were the key factors for methods validations.
This paper presents morphological operators with non-fixed shape kernels, or amoebas, which take into account the image contour variations to adapt their shape. Experiments on grayscale and color ...images demonstrate that these novel filters outperform classical morphological operations with a fixed, space-invariant structuring element for noise reduction applications. Tests on synthetic 3D images are then performed to show the high noise-reduction capacity of amoeba-based filters.
Purpose
This paper aims to discuss the scanning methodology depending on the close-range photogrammetry technique, which is appropriate for the precise three-dimensional (3D) modelling of objects in ...millimetres, such as the dimensions and structures in sub-millimetre scale.
Design/methodology/approach
The camera was adjusted to be tilted around the horizontal axis, while coded dot targets were used to calibrate the digital camera. The experiment was repeated with different rotation angles (5°, 10°, 15°, 20°, 25°, 30°, 50° and 60°). The images were processed with the PhotoModeler software to create the 3D model of the sample and estimate its dimensions. The features of the sample were measured using high-resolution transmission electron microscopy, which has been considered as a reference and the comparative dimensions.
Findings
The results from the current study concluded that changing the rotation angle does not significantly affect the results, unless the angle of imagery is large which prevent achieving about 20: 30% overlap between the images but, the more angle decreases, the more number of images increase as well as the processing duration in the programme.
Originality/value
Develop an automatic appropriate for the precise 3D modelling of objects in millimetres, such as the dimensions and structures in sub-millimetre scale using photogrammetry.
Measuring the similarity between point clouds is required in many areas. In autonomous driving, point clouds for 3D perception are estimated from camera images but these estimations are error-prone. ...Furthermore, there is a lack of measures for quality quantification using ground truth. In this paper, we derive conditions point cloud comparisons need to fulfill and accordingly evaluate the Chamfer distance, a lower bound of the Gromov Wasserstein metric, and the ratio measure. We show that the ratio measure is not affected by erroneous points and therefore introduce the new measure “average ratio”. All measures are evaluated and compared using exemplary point clouds. We discuss characteristics, advantages and drawbacks with respect to interpretability, noise resistance, environmental representation, and computation.
Face is the most preferred biometric trait used to recognize a person. The 2D face can be considered as a promising biometric trait; however, it may be affected by changes in the age, skin color, ...texture, or any other environmental factors like illumination variations, occlusion, and low image resolution. The 3D face is an emerging biometric trait, which is being used recently for human recognition. The discriminating power of the 3D face is highly motivating for many tasks such as security, surveillance, and many other technological application in day to day life. Although there are many techniques available for 3D face recognition, most of these techniques are based on volumetric or depth/range images. The conversion of 3D face data, which is originally in point cloud format to volumetric representation makes the data bulkier. Further, some of the geometric properties may be lost when 3D data is converted to a representation of lower dimensions such as depth/range images. The driving objective behind this research is to perform 3D face recognition by directly using faces represented in the form of 3D point cloud. We propose a novel approach for 3D face recognition by learning the similarity and dissimilarity in 3D faces, and for this purpose, introduce a triplet network. The network is an ensemble of our proposed Convolutional PointNet (CPN) network, used for feature extraction and triplet loss. The proposed network maps a 3D face data to Euclidean space where distance based scores represent the similarity among the 3D faces. We also introduce a new evaluation approach for computing the dissimilarity between highly similar 3D face biometric data. Experimentation has been carried out on two databases, namely IIT Indore 3D face database (our in-house database) and Bosphorus 3D face database. To handle the training issues due to the limited availability of samples for each subject in both the databases, we propose a technique for 3D data augmentation. We perform various experiments using the proposed network and show the performance in terms of verification rate and ROC curve. Our point cloud based triplet network shows encouraging performance as compared to other state-of-the-art techniques.
PDF
