一种场景自适应的双分支牛脸高效识别算法

A Scene-Adaptive Dual-Branch Efficient Cattle Facial Recognition Algorithm

随着智慧牧业的高速发展,牛脸识别已成为牛场智能化养殖的关键,但牛场养殖环境复杂且动物的自主能动性差,导致牛脸数据采集与识别过程会受到模糊、遮挡和光照等环境因素的严重干扰.针对此问题,提出一种复杂场景自适应选择双分支牛脸高效识别算法.该算法首先设计了基于像素融合的数据增强策略,通过Beta分布计算融合系数,将牛的左右脸图像按融合系数进行像素级整合,在丰富样本特征信息同时,增强网络学习模糊和遮挡下的牛脸特征,提升网络对复杂场景的泛化能力;其次,在主干特征提取网络中引入一种新型注意力机制CDAA(Composite Dual-branch Adaptive Attention),可随着场景信息变换,自适应加强通道与空间注意力分支的权重,提高网络在复杂场景下的特征筛选能力;之后,设计FaceNet与U-LBP(Uniform Local Binary Patterns)结合的双分支特征提取结构,并将提取的特征向量实现自适应加权融合,增加网络在过亮或过暗环境下的鲁棒性;最后,在损失函数中加入改进交叉熵损失(Focal Loss),根据场景信息复杂度动态调控权重系数,实现对难易分类样本自主控制.为检测算法的有效性和实时性,在特定数据集上进行消融试验,与多种典型识别算法进行对比.实结果表明,提出的算法能很好满足实时性要求,在开集测试集上准确率达到87.53%,识别速度达到108帧/s,且在复杂场景下,识别效果均优于对比网络.

With the rapid development of intelligent animal husbandry,cattle facial recognition has become a key aspect of intelligent farming in cattle ranches.However,due to the complexity of the ranching environment and the limited autonomy of animals,the process of collecting and identifying cattle facial data is severely affected by environmental factors such as blur⁃riness,occlusion,and lighting.To address this issue,a complex scene-adaptive dual-branch efficient cattle facial recognition al⁃gorithm is proposed.This algorithm first designs a data augmentation strategy based on pixel fusion.By calculating fusion co⁃efficients using the beta distribution,the left and right facial images of cattle are integrated at the pixel level,enriching the sam⁃ple's feature information.Simultaneously,the algorithm enhances the network's ability to learn cattle facial features under blurriness and occlusion,improving its generalization ability to complex scenes.Furthermore,a novel attention mechanism called composite dual-branch adaptive attention(CDAA)is introduced into the main feature extraction network.This mecha⁃nism adaptively strengthens the weights of the channel and spatial attention branches as scene information changes,enhancing the network's feature selection ability in complex scenarios.Next,a dual-branch feature extraction structure combining FaceNet and U-LBP(Uniform Local Binary Patterns)is designed.The extracted feature vectors are adaptively weighted and fused to increase the network's robustness in overly bright or dark environments.Finally,an improved cross-entropy loss(Fo⁃cal Loss)is incorporated into the loss function.Weight coefficients are dynamically adjusted based on the complexity of the scene information to autonomously control the classification of difficult and easy samples.To evaluate the effectiveness and re⁃al-time performance of the algorithm,ablation experiments are conducted on a specific dataset,comparing it with various typi⁃cal recognition algorithms.The experimental results indicate that the proposed algorithm effectively meets real-time require⁃ments,achieving an accuracy of 87.53%on the open test set with a recognition speed of 108 frames per second.Moreover,in complex scenarios,the recognition performance of the proposed algorithm surpasses that of the comparative networks.