Maintaining the health and well-being of animals is critical to the efficiency and profitability of livestock operations. However, it can be difficult to monitor the health of animals in large group-housed settings without the assistance of technology. This study presents a system that uses depth images to continuously track individual pigs in a group-housed environment. It is an alternative to traditional manual observation used by both researchers and producers for the analysis of animal activities and behaviours. The tracking method used by the system exploits the consistent shape and fixed number of the targets in the environment by applying expectation maximisation as a policy for fitting an ellipsoid to each target. Results demonstrate that the system can maintain the correct positions and orientations of 15 group-housed pigs for an average of 19.7 min between failure events.

The development of vessels can provide important information about the growth status of animal embryos. It is, therefore, important to automatically locate the deformed vessel branches from the embryo images. However, very few vessel detectors can accurately locate all vessel branches when the captured images are low quality and the implied vessel shapes are complex. In this study, a new framework consisting of vessel region extraction and snake shape optimisation is proposed. The main contribution in this detector is a novel open snake model based on the global guidance field and deformation template initialisation. Experimental results on a specific application of an embryo vessel database [Database and source codes: https://github.com/wcxie/Egg-embryro-vessel-location/.] demonstrate that the proposed algorithm not only locates the vessel shape properly but also obtains the orientations of embryo vessel branches accurately. Comparison to traditional guidance fields and the active appearance model illustrates the effectiveness and competitiveness of the proposed model.

Insect detection is one of the most challenging problems of biometric image processing. This study focuses on developing a method to detect both individual insects and touching insects from trap images in extreme conditions. This method is able to combine recent approaches on contour-based and region-based segmentation. More precisely, the two contributions are: an adaptive k-means clustering approach by using the contour's convex hull and a new region merging algorithm. Quantitative evaluations show that the proposed method can detect insects with higher accuracy than that of the most used approaches.

In order to monitor an animal population and to track individual animals in a non-invasive way, identification of individual animals based on certain distinctive characteristics is necessary. In this study, automatic image-based individual identification of the endangered Saimaa ringed seal (Phoca hispida saimensis) is considered. Ringed seals have a distinctive permanent pelage pattern that is unique to each individual. This can be used as a basis for the identification process. The authors propose a framework that starts with segmentation of the seal from the background and proceeds to various post-processing steps to make the pelage pattern more visible and the identification easier. Finally, two existing species independent individual identification methods are compared with a challenging data set of Saimaa ringed seal images. The results show that the segmentation and proposed post-processing steps increase the identification performance.

Facial expressions of laboratory mice provide important information for pain assessment to explore the effect of drugs being developed for medical purposes. For automatic pain assessment, a mouse face tracker is needed to extract the face regions in videos recorded in pain experiments. However, since the body and face of mice are the same colour and mice move fast, tracking their face is a challenging task. In recent years, with their ability to learn from data, deep learning provides effective solutions for a wide variety of problems. In particular, convolutional neural networks (CNNs) are very successful in computer vision tasks. In this study, a CNN based tracker network called MFTN is proposed for mouse face tracking. CNNs are good at extracting hierarchical features from the training dataset. High-level features contain semantic features and low-level features have high spatial resolution. In the proposed MFTN architecture, target information is extracted from a combination of low- and high-level features by a sub-network, namely the Feature Adaptation Network (FAN), to achieve a robust and accurate tracker. Among the MFTN versions, the MFTN/c tracker achieved an accuracy of 0.8, robustness of 0.67, and a throughput of 213 fps on a workstation with GPU.

Extracting a statistically significant result from video of natural phenomenon can be difficult for two reasons: (i) there can be considerable natural variation in the observed behaviour and (ii) computer vision algorithms applied to natural phenomena may not perform correctly on a significant number of samples. This study presents one approach to clean a large noisy visual tracking dataset to allow extracting statistically sound results from the image data. In particular, analyses of 3.6 million underwater trajectories of a fish with the water temperature at the time of acquisition are presented. Although there are many false detections and incorrect trajectory assignments, by a combination of data binning and robust estimation methods, reliable evidence for an increase in fish speed as water temperature increases are demonstrated. Then, a method for data cleaning which removes outliers arising from false detections and incorrect trajectory assignments using a deep learning-based clustering algorithm is proposed. The corresponding results show a rise in fish speed as temperature goes up. Several statistical tests applied to both cleaned and not-cleaned data confirm that both results are statistically significant and show an increasing trend. However, the latter approach also generates a cleaner dataset suitable for other analysis.

In the field of applied animal behaviour, video recordings of a scene of interest are often made and then evaluated by experts. This evaluation is based on different criteria (number of animals present, an occurrence of certain interactions, the proximity between animals and so forth) and aims to filter out video sequences that contain irrelevant information. However, such task requires a tremendous amount of time and resources, making manual approach ineffective. To reduce the amount of time the experts spend on watching the uninteresting video, this study introduces an automated watchdog system that can discard some of the recorded video material based on user-defined criteria. A pilot study on cows was made where a convolutional neural network detector was used to detect and count the number of cows in the scene as well as include distances and interactions between cows as filtering criteria. This approach removed 38% (50% for additional filter parameters) of the recordings while only losing 1% (4%) of the potentially interesting video frames.

Image identification of animals is mostly centred on identifying them based on their appearance, but there are other ways images can be used to identify animals, including by representing the sounds they make with images. In this study, the authors present a novel and effective approach for automated identification of birds and whales using some of the best texture descriptors in the computer vision literature. The visual features of sounds are built starting from the audio file and are taken from images constructed from different spectrograms and from harmonic and percussion images. These images are divided into sub-windows from which sets of texture descriptors are extracted. The experiments reported in this study using a dataset of Bird vocalisations targeted for species recognition and a dataset of right whale calls targeted for whale detection (as well as three well-known benchmarks for music genre classification) demonstrate that the fusion of different texture features enhances performance. The experiments also demonstrate that the fusion of different texture features with audio features is not only comparable with existing audio signal approaches but also statistically improves some of the stand-alone audio features. The code for the experiments will be publicly available at https://www.dropbox.com/s/bguw035yrqz0pwp/ElencoCode.docx?dl=0.

Melanoma is one the most increasing cancers since past decades. For accurate detection and classification, discriminative features are required to distinguish between benign and malignant cases. In this study, the authors introduce a fusion of structural and textural features from two descriptors. The structural features are extracted from wavelet and curvelet transforms, whereas the textural features are extracted from different variants of local binary pattern operator. The proposed method is implemented on 200 images from dermoscopy database including 160 non-melanoma and 40 melanoma images, where a rigorous statistical analysis for the database is performed. Using support vector machine (SVM) classifier with random sampling cross-validation method between the three cases of skin lesions given in the database, the validated results showed a very encouraging performance with a sensitivity of 78.93%, a specificity of 93.25% and an accuracy of 86.07%. The proposed approach outperforms the existing methods on the database.

Visual target tracking has long been a challenging problem because of the variable appearance of the target with changing spatiotemporal factors. Therefore, it is important to design an effective and efficient appearance model for tracking tasks. This study proposes a tracking algorithm based on locality-constrained linear coding (LLC) under a particle filtering framework. A local feature descriptor is presented that can evenly represent the local information of each patch in the tracking region. LLC uses the locality constraints to project each local feature descriptor into its local-coordinate system. Compared with sparse coding, LLC can be performed very quickly for appearance modelling because it has an analytical solution derived by a three-step matrix calculation, and the computational complexity of the proposed tracking algorithm is . Both quantitative and qualitative experimental results demonstrate that the authors’ proposed algorithm performs favourably against the 10 state-of-the-art trackers on 12 challenging test sequences. However, related experimental results show that the performance of their tracker is not effective enough for small tracking targets owing to a lack of sufficient local region information.

Remote sensing images taken in hazy situations are degraded by scattering of atmospheric particles, which greatly influences the efficiency of visual systems. Therefore, the visibility restoration of hazy images becomes a significant area of research. In this study, a fourth-order partial differential equations based trilateral filter (FPDETF) dehazing approach is proposed to enhance the coarse estimated atmospheric veil. FPDETF is able to reduce halo and gradient reversal artefacts. It also preserves the radiometric information of haze-free images. The visibility restoration phase is also refined to reduce the colour distortion of dehazed images. The proposed technique has been evaluated on ten well-known remote sensing images and also compared with seven well-known existing dehazing approaches. The experimental results reveal that the proposed technique outperforms others in terms of contrast gain and percentage of saturated pixels.

This study focuses on fast and robust outlier matches removal strategy to improve the efficiency and precision of initial alignment and further the quality of pairwise registration. Starts from the point matches obtained via feature detecting and matching, the distance disparity matrix derived from Euclidean invariants of rigid transformation is introduced, based on which a fast and effective pruning method is proposed to eliminate the outlier correspondences, especially the sharp ones. Then, the remaining matches are sent into the enhanced least-square backward method to estimate an initial transformation in lesser attempts. Since most of the outliers are rejected, presented backward method could provide a finer alignment to input point clouds in higher efficiency than existing methods, and the following refining procedure converges to a more precise registration consuming fewer iterations, which have been proved in designed experiments. The thresholds employed in the pipeline are all automatically determined according to the actual resolution of input point clouds. Users are just required to control the error precision through a scale factor, in which way the inaccuracy and inconvenience of manually threshold defining are avoided.