The quality of uni-directional tape in its production process is affected by environmental conditions like temperature and production speed. In this paper, computer vision algorithms on the scanned images are needed to be used in this context to detect and classify tape damages during the manufacturing procedure. We perform a comparative study among famous feature descriptors for fault candidate generation, then propose own features for fault detection. We investigate various machine learning techniques to find best model for the classification problem. The empirical results demonstrate the high performance of the proposed system and show preference of random forest and canny edges for classifier and feature generator respectively.

Object tracking is used to locate the position of an object over a period of time using the association of an object of interest over consecutive frames. In the last years, several methods were proposed to track objects in rectangular videos. This paper presents is an object tracking method within 360-degree videos using a state-of-the-art tracking-by-detection paradigm. This method uses two trackers namely Kalman filter and Lucas-Kanade methods to handle challenges in the 360-degree videos. The proposed method uses a deep learning object detector for extraction of prior information of the object of interest. The information is then used, to track the object of interest using a combination of the two trackers of the Kalman filter and Lucas-Kanade. The experiments show that this combination improves the tracker stability.

The task of fast object tracking in polar images using emerging high-resolution 360-degree camera technology is presented in this paper. In this approach, when an arbitrary object has been selected in the first frame, the proposed method searches for the object in the next frames. This task is challenging when the video contains complexity which cannot be handled by common tracking methods. The main contribution of this paper uses polar object selection and color binary features to facilitate robust object tracking in 360-degree images. Using the proposed polar object selection method, each object is represented by a polar component and high performance of the tracking algorithm in terms of precision and speed is achieved. We evaluate the applicability of our approach on a new dataset containing more than 30000 frames of 360-degree images wherein high performance in challenging real-world scenarios is demonstrated. The proposed algorithm outperforms the related methods.

A vision-based method for detecting the cracks in the concrete sleepers of the railway tracks will be introduced in this paper. The method is able to detect and partially classify the cracks of the concrete sleepers in two successive steps based on the image processing and pattern recognition techniques. The method has been implemented on the acquired image data frames followed by the analysis, experimental, comparison results and evaluation. The presented results are reasonable which indicates the goodness of the introduced method. The preliminary results of this work have been presented in [A. Delforouzi, A. H. Tabatabaei, M. H. Khan and M. Grzegorzek, A vision-based method for automatic crack detection in railway sleepers, in Kurzynski, M., Wozniak, M., Burduk, R. (eds.), Proceedings of the 10th International Conference on Computer Recognition Systems CORES 2017, Polanica Zdroj, Poland. CORES 2017. Advances in Intelligent Systems and Computing, Vol. 578 (Springer, Cham, 2018), pp. 130–139, doi: 10.1007/978-3-319-59162-9_14].

Object tracking in challenging videos is a hot topic in machine vision. Recently, novel training-based detectors, especially using the powerful deep learning schemes, have been proposed to detect objects in still images. However, there is still a semantic gap between the object detectors and higher level applications like object tracking in videos. This paper presents a comparative study of outstanding learning-based object detectors such as ACF, Region-Based Convolutional Neural Network (RCNN), FastRCNN, FasterRCNN and You Only Look Once (YOLO) for object tracking. We use an online and offline training method for tracking. The online tracker trains the detectors with a generated synthetic set of images from the object of interest in the first frame. Then, the detectors detect the objects of interest in the next frames. The detector is updated online by using the detected objects from the last frames of the video. The offline tracker uses the detector for object detection in still images and then a tracker based on Kalman filter associates the objects among video frames. Our research is performed on a TLD dataset which contains challenging situations for tracking. Source codes and implementation details for the trackers are published to make both the reproduction of the results reported in this paper and the re-use and further development of the trackers for other researchers. The results demonstrate that ACF and YOLO trackers show more stability than the other trackers.

In this paper, a method for automatic selection and classification of the sleeper cracks is presented. This method includes three main sequential steps of image pre-processing, sleeper detection and crack detection. Two approaches including rule-based method and template matching method in the frequency domain are proposed for the sleeper detection step. We utilize adaptive threshold binarization to handle challenging crack detection under non-uniform lightening condition and hierarchical structure for the decision making step. Two unsupervised classifiers are exploited to detect the cracks. The results show that the presented method has the overall detection rate with accuracy of at least 87 percent.

The task of object tracking in rectangular videos has been addressed in recent years by many researchers, where each method tries to propose a solution for a special challenge. Handling a variety of challenging situation of object tracking in 360-degree videos is still an unsolved problem and needs to be more considered. In the real world, the challenging situations include moving camera, high-resolution videos, background clutter, fast processing, in and out-of-plane rotation, small, articular and rigid objects, illumination variation and partial and full occlusion. In this paper, a new structure of SURF-based object tracking is proposed which uses a train-based matching to address the challenging object tracking in 360-degree videos. The proposed tracker is able to estimate out-of-plane rotation and occlusion during the tracking and adapt itself to handle it. Our experiments demonstrate the robustness of our tracker.

In this paper, a method for unknown object tracking in output images from 360-degree cameras called Modified Training-Learning-Detection (MTLD) is presented. The proposed method is based on the recently introduced Training-Learning-Detection (TLD) scheme in the literature. The flaws of the TLD approach have been detected and significant modifications are proposed to enhance and to elaborate the scheme. Unlike TLD, MTLD is capable of detecting the unknown objects of interest in 360-degree images. According to the experimental results, the proposed method significantly outperforms the TLD method in terms of detection rate and implementation cost.

this paper presents a practical Farsi license plate recognition system that can process images at high accuracy rates for real-time applications. This system is designed to exhibit the requirements of performance, computational speed, for vehicle surveillance applications, such as stolen car detection systems and traffic monitoring. These requirements are satisfied by adopting a hierarchy serial framework. The proposed system is composed of five main cascading parts for motion detection, plate location, character segmentation, character recognition, and postprocessing. In each part many candidates are selected and some of them are rejected from desired candidates list and other forward to next step. This strategy saves computational load and increase implementation velocity. Experimental results show that the proposed system can provide over 94% accuracy with a resolution of 576*768 pixels on a 3GHz Intel Pentium 4 and 1GB RAM personal computer.

In this paper a novel Audio steganography method is proposed. In this algorithm, first embedding threshold in the time domain is estimated. Then this threshold is used for data concealment in the time domain. The performance of the proposed scheme in terms of SNR (signal to noise ratio) and BER (bit error rate) is analyzed. An estimation formula that connects SNR with embedding threshold has been provided to ensure the transparency of embedded data. The security of this method is enhanced by using of an encryption method prior to the data embedding step. Proposed scheme is a high velocity method because of its implementation in the temporal domain. Experimental results demonstrate that the proposed method has high payload, full and blind recovery and high audio quality.

In this paper a novel method for digital audio steganography is presented where encrypted covert data is embedded into the coefficients of the host audio (cover signal) in the integer wavelet domain. The hearing threshold is calculated in the integer domain and this threshold is employed as the embedding threshold. The inverse integer wavelet transform is applied to the modified coefficients to form a new audio sequence (stego signal). The characteristics of this method are large payload, high audio quality and full recovery.

This paper presents a robust method of audio watermarking in the wavelet domain. In this method, the watermarked data is encrypted, then combined with a synchronization code and embedded in low- frequency coefficients of wavelet transform. In this paper two techniques of quantization are employed for data embedding process. The magnitude of quantization step and embedding strength is adaptively determined according to the characteristics of the human auditory system (HAS). Therefore data embedding distortion is imperceptible for human ears. The experimental results show that the proposed watermarking scheme is transparence and very robust against common attacks such as additive Gaussian noise, low-pass filtering, resampling, requantization, shifting, cropping and MPEG compression.

In this paper we present a novel method for digital audio steganography where encrypted covert data is embedded into the wavelet coefficients of host audio signal. To avoid extraction error we use lifting wavelet transform. For using the maximum capacity of audio signals, we calculate hearing threshold in wavelet domain. Then according to this threshold data bits are embedded in the least significant bits of lifting wavelet coefficients. Inverse lifting wavelet transform is applied to modified coefficients to construct stego signal in time domain. Experimental results show that proposed method has large payload, high audio quality and full recovery.

This paper presents a novel audio watermarking scheme based on echo hiding technique. This method is different from conventional echo hiding methods which embed only one bit in each audio section. In this method by using different echo delays we can embed more than one bit in each audio sections. For enhancing robustness of proposed algorithm against various types of attacks, we exploit dual echo kernels. Experimental results show that proposed method has good performance in robustness and inaudibility point of view.