Search

Highlights • Aggressive pigs were separated from the herd and extracted continuously. • Head and 4 kink points of pigs were located and used as feature points. • Kinetic energy model was built by analysing motion of these 5 feature points. • Kinetic energy difference between adjacent frames was used to recognise aggression. • The proposed algorithm has higher accuracy, sensitivity, specificity and precision. Abstract The aim of this study is to build a kinetic energy model to recognise aggressive behaviours in pigs. Pig mixing experiments were performed 10 times. In each experiment, 7 pigs from 3 pens were mixed and then videos of the first 3 h after mixing and the 3 h after 24 h were selected as experimental data. A 30 h data of 5 groups of pigs were used as training set and a 30 h data of another 5 groups of pigs as test set. Firstly, according to connected area and adhesion index, aggressive pigs were separated from the herd and then key frame sequences with aggression were extracted. Secondly, head and 4 kink points of pigs were selected as feature points. By analysing motion of these points, kinetic energy of 2 aggressive pigs was calculated and kinetic energy difference between adjacent frames was selected as features. Finally, these features were trained by using hierarchical clustering to obtain the thresholds of high and medium aggression. These thresholds were used to design the rule of aggression recognition. By using the proposed algorithm, high aggression could be recognised with an accuracy of 95.8%, a sensitivity of 92.8%, specificity of 96.7% and precision of 90.2%, and medium aggression could be recognised with an accuracy of 92.3%, a sensitivity of 92.6%, specificity of 91.9% and precision of 93.0%. The results indicate that this algorithm can be used to recognise aggressive behaviours of pigs. [ABSTRACT FROM AUTHOR]

The aim of this study is to develop a computer vision-based method to automatically detect aggressive behaviors among pigs. Ten repetitions of the same experiment were performed. In each of the experiment, 7 piglets were mixed from three litters in two pigpens and captured on video for a total of 6 h. From these videos, the first 3 h of video after mixing were recorded as a training set, and the 3 h of video after 24 h were recorded as a validation set. Connected area and adhesion index were used to locate aggressive pigs and to extract key frame sequences. The two pigs in aggression were regarded as a whole rectangle according to their characteristics of continuous and large-proportion adhesion. The acceleration feature was extracted by analyzing the displacement change of four sides of this rectangle between adjacent frames, and hierarchical clustering was used to calculate its threshold. Based on this feature, the recognition rules of medium and high aggression were designed. Testing 10 groups of pigs, the accuracy of recognizing medium aggression was 95.82% with a sensitivity of 90.57% and with a specificity of 96.95%, and the accuracy of recognizing high aggression was 97.04% with a sensitivity of 92.54% and with a specificity of 97.38%. These results indicate that the acceleration can be used to recognize pigs’ aggressive behaviors. [ABSTRACT FROM AUTHOR]

This paper proposes a method to segment apples on trees at night for apple-harvesting robots based on color and position of pixels. Images of apples acquired under artificial light with low illumination at night include less color information than daytime images, so it is necessary to take position of pixels into consideration. The new method has two main steps. Firstly, color components of sampled pixels in RGB and HSI color space are used to train a neural network model to segment the apples. However, the segmentation results are incomplete and not able to guide apple-harvesting robots accurately, because partial edge regions of apples are dark in shadows and difficult to be recognized due to uneven illumination. Secondly, the color and position of pixels around segmented regions and pixels on the boundary of segmented regions are taken into consideration to segment the edge regions of apples. The union of two segmentation results is the final result. The complete recognition can increase the accuracy of location by about 6.5%, which verified the validity and feasibility of the method. [ABSTRACT FROM AUTHOR]

• The proposed model achieves 85.3% segmentation accuracy for green fruits. • FPN is replaced by RFPN, which improves the segmentation accuracy. • A double-layer CBA network is added to segment overlapping fruits more accurately. Accurate detection and segmentation of obscured green fruits is a key factor in the efficiency of the harvesting robot. In the real and complex orchard environment, there are many influencing factors such as illumination changes, occlusion of branches and leaves, weather, etc., which bring difficulties to the work of picking robots. Therefore, higher requirements are put forward for accurate detection and segmentation algorithms of machine vision. An accurate detection and segmentation method for obscured green fruit based on a fully convolutional one-stage (FCOS) object detection model is proposed. First, the feature pyramid network (FPN) in FCOS is replaced by residual feature pyramid network (RFPN), which improves the detection accuracy of green fruits with different sizes. Next, as the boundaries of fuzzy fruits are often not accurately identified, a two-layer convolutional block attention network is connected to a modified FCOS to recover the edges of incomplete green fruits. The experimental results show that the detection and segmentation accuracy of green fruit reaches 81.2% and 85.3% on Apple dataset. On the Apple-ape dataset, the detection accuracy and segmentation accuracy reached 77.2% and 79.7%, respectively. The Params of the model is 39.7 M, and the floating point operations (FLOPs) are 169.8G, which has great advantages over other ten advanced models in terms of accuracy and complexity. The model also improves the working efficiency of orchard harvesting robots and lays the foundation for the further development of intelligent agriculture. [ABSTRACT FROM AUTHOR]

• Adding identity mapping and channel attention into one-shot aggregation units. • Predict coarse instance-aware bounding boxes utilizing pixel-by-pixel regression. • A spatial attention-guided mask was used to focus on information pixels and suppress irrelevant pixels. • Perform fine mask segmentation within the detected and attention-guided coarse bounding box. • Small-sized, efficient, and suitable for integrated applications of automatic feeding boats. Effective biomass detection methods utilizing computer vision techniques should be capable of handling large differences in object posture and various degrees of mutual overlap/occlusion in object scenes. We propose an automatic coarse-to-fine joint detection and instance segmentation network (JDSNet) that can perform real-time detection and instance segmentation on underwater non-structural live crabs in real-time. The method was adapted to non-structural objects and mobile devices by applying the anchor-free mechanism and center-ness strategy of the fully convolutional one-stage prediction head and the improved energy-efficient backbone IVoVNet-19-DW with identity mapping and channel attention. This approach avoided the complicated computations related to the anchor-based mechanism and effectively generated the features of various receiving domains, jointly improving the memory access speed and accuracy of predicting the bounding boxes of different instances. A novel spatial attention-guided mask branch was then added to focus on irregular occluded object pixels and conduct precise pixel-level mask segmentation within the predicted coarse instance-aware rectangular-bounding boxes. The experimental analysis using the proposed method resulted in a quality of detection F 1 and segmentation D ic of 97.7% and 94.6%, respectively. The fastest detection speed of a single image was 48.07/13.32 fps (~10 times faster than the existing network Mask RCNN) on a commonly configured GPU/CPU, requiring only 7.04 MB of storage (~25 times smaller than Mask RCNN). It indicates that JDSNet can segment various non-structural live crabs and perform biomass statistics robustly and efficiently, exhibiting significance for precision feeding applications in automatic feeding boats. [ABSTRACT FROM AUTHOR]