1. |
- Ardö, Hakan, et al.
(author)
-
Convolutional neural network-based cow interaction watchdog
- 2018
-
In: IET Computer Vision. - : Institution of Engineering and Technology (IET). - 1751-9632 .- 1751-9640. ; 12:2, s. 171-177
-
Journal article (peer-reviewed)abstract
- 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.
|
|
2. |
- Guzhva, Oleksiy
(author)
-
Computer vision algorithms as a modern tool for behavioural analysis in dairy cattle
- 2018
-
Doctoral thesis (other academic/artistic)abstract
- Looking at modern dairy production, loose housing, i.e. free stalls became one of the most common practices, which, while widely implemented along with different management routines, do not always include the adjustments necessary for assuring animal welfare. The analysis of interactions occurring between cows in dairy barns and their effect on health and performance is of great importance for sustainable, animal-friendly production. The general aim of this thesis was to investigate the possibilities and limitations of computer vision approach for studying dairy cattle behaviour and interactions between animals, as well as take a first step towards the fully automated system for continuous surveillance in modern dairy barns. In the first study, a seven-point shape-model for describing a cow from the mathematical perspective was proposed and investigated. A pilot study showed that the proposed Behavioural Detector based on the developed shape-model provided a solid basis for behavioural studies in a real-life dairy barn environment. The second study investigated a classification case from the industry: how animal distribution and claw positioning in specific areas could affect the maximal load on floor elements. The results of the study provided more substantial background data for determining the dimensioning of the strength of the slats. The third study aimed to take the first step towards an automated system (so-called WatchDog) for behavioural analysis and automatic filtering of the recorded video material. The results showed that the proposed solution is capable of detecting potentially interesting scenes in video-material with the precision of 92,8%. In the fourth and final study, a state-of-the-art tracking/identification algorithm for multiple objects with near-real-time implementation in crowded scenes with varying illumination was developed and evaluated. The algorithms forming the multi-modular WatchDog system and developed during this project are the crucial stepping stone towards a fully-automated solution for continuous surveillance of health and welfare-related parameters in dairy cattle. The proposed system could also serve as evaluation/benchmark tool for modern dairy barn assessment. Keywords: dairy cattle, image analysis, Precision Livestock Farming, computer vision, deep learning, convolutional neural networks, social interactions, tracking, cow traffic
|
|
3. |
- Guzhva, Oleksiy, et al.
(author)
-
Now you see me : Convolutional neural network based tracker for dairy cows
- 2018
-
In: Frontiers in robotics and AI. - : Frontiers Media SA. - 2296-9144. ; 5:SEP
-
Journal article (peer-reviewed)abstract
- To maintain dairy cattle health and welfare at commensurable levels, analysis of the behaviors occurring between cows should be performed. This type of behavioral analysis is highly dependent on reliable and robust tracking of individuals, for it to be viable and applicable on-site. In this article, we introduce a novel method for continuous tracking and data-marker based identification of individual cows based on convolutional neural networks (CNNs). The methodology for data acquisition and overall implementation of tracking/identification is described. The Region of Interest (ROI) for the recordings was limited to a waiting area with free entrances to four automatic milking stations and a total size of 6 × 18 meters. There were 252 Swedish Holstein cows during the time of study that had access to the waiting area at a conventional dairy barn with varying conditions and illumination. Three Axis M3006-V cameras placed in the ceiling at 3.6 meters height and providing top-down view were used for recordings. The total amount of video data collected was 4 months, containing 500 million frames. To evaluate the system two 1-h recordings were chosen. The exit time and gate-id found by the tracker for each cow were compared with the exit times produced by the gates. In total there were 26 tracks considered, and 23 were correctly tracked. Given those 26 starting points, the tracker was able to maintain the correct position in a total of 101.29 min or 225 s in average per starting point/individual cow. Experiments indicate that a cow could be tracked close to 4 min before failure cases emerge and that cows could be successfully tracked for over 20 min in mildly-crowded ( < 10 cows) scenes. The proposed system is a crucial stepping stone toward a fully automated tool for continuous monitoring of cows and their interactions with other individuals and the farm-building environment.
|
|