Lightweight individual cow identification based on Ghost combined with attention mechanism

Abstract

Individual cow identification is a prerequisite for intelligent dairy farming management, and is important for achieving accurate and informative dairy farming. Computer vision-based approaches are widely considered because of their non-contact and practical advantages. In this study, a method based on the combination of Ghost and attention mechanism is proposed to improve ReNet50 to achieve non-contact individual recognition of cows. In the model, coarse-grained features of cows are extracted using a large sensory field of cavity convolution, while reducing the number of model parameters to some extent. ResNet50 consists of two Bottlenecks with different structures, and a plug-and-play Ghost module is inserted between the two Bottlenecks to reduce the number of parameters and computation of the model using common linear operations without reducing the feature map. In addition, the convolutional block attention module (CBAM) is introduced after each stage of the model to help the model to give different weights to each part of the input and extract the more critical and important information. In our experiments, a total of 13 cows’ side view images were collected to train the model, and the final recognition accuracy of the model was 98.58%, which was 4.8 percentage points better than the recognition accuracy of the original ResNet50, the number of model parameters was reduced by 24.85 times, and the model size was only 3.61 MB. In addition, to verify the validity of the model, it is compared with other networks and the results show that our model has good robustness. This research overcomes the shortcomings of traditional recognition methods that require human extraction of features, and provides theoretical references for further animal recognition.

A review of on-farm recording tools for smallholder dairy farming in developing countries

“…Accurate individual cow identification is needed to achieve accuracy and provide valuable information in dairy farming. Individual records can enhance farm management decisions at the individual cow and herd levels, aiding in genetic evaluation, selection, and management choices (Fu et al 2022;Lassen et al 2023). Farmers can optimize milk production by modifying each cow's diet and nutrition using milk production data.…”

Open-Set Recognition of Individual Cows Based on Spatial Feature Transformation and Metric Learning

“…The feature extraction network architecture is shown in Figure 6. [21,24,25] and are considered adequate. ResNet101 is a variant of the ResNet [31] architecture that contains 101 layers.…”

“…We selected well-known feature extraction networks, including MobileNet V2, ResNet101, ResNeSt101, and Vision Transformer (ViT), as baselines. These networks, encompassing lightweight, residual, and Transformer visual networks and their variants, have been proven effective in the field of individual animal identification [17,21]. In this study, we focus on the model's performance in processing images of cows from an overhead perspective, characterized by complex backgrounds and variable postures.…”

“…Other researchers have also proposed various methods for cattle facial recognition [17,18]. Additionally, some studies focus on identifying individuals based on other body parts of cattle [19,20] and on model lightweighting [21][22][23]. However, whether based on the cow's back, face, or other body parts, most studies on individual dairy cow recognition focus on closed-set identification, which only recognizes individuals within the training set.…”

A novel vibro-acoustic fault diagnosis approach of planetary gearbox using intrinsic wavelet integrated GE-EfficientNet

“…First, Ghost uses standard convolution operation to generate part of the original feature map. The resulting original features are then subjected to low-cost linear transformations to enhance the features and increase channels [33,34]. Finally, the two sets of features are spliced together.…”