Environmental noise can pose a threat to the stable operation of current speech recognition systems. It is therefore essential to develop a front feature set that is able to identify speech under low signal-to-noise ratio. In this paper, a robust fusion feature is proposed that can fully characterize speech information. To obtain the cochlear filter cepstral coefficients (CFCC), a novel feature is first extracted by the power-law nonlinear function, which can simulate the auditory characteristics of the human ear. Speech enhancement technology is then introduced into the front end of feature extraction, and the extracted feature and their first order difference is combined in new mixed features. An energy feature Teager energy operator cepstral coefficient (TEOCC) is also extracted, and combined with the above mixed features to form fusion feature sets. Principal component analysis (PCA) is then applied to feature selection and optimization of the feature set, and the final feature set is used in a non-specific persons, isolated words, and small-vocabulary speech recognition system. Finally, a comparative experiment of speech recognition is designed to verify the advantages of the proposed feature set using support vector machine (SVM). The experimental results show that the proposed feature set not only display a high recognition rate and excellent anti-noise performance in speech recognition, but can also fully characterize the auditory and energy information in speech signals.

Classifying and accurately locating a visual category with few annotated training samples in computer vision has motivated the few-shot object detection technique, which exploits transfering the source-domain detection model to the target domain. Under this paradigm, however, such transferred source-domain detection model usually encounters difficulty in the classification of the target domain because of the low data diversity of novel training samples. To combat this, we present a simple yet effective few-shot detector, Transferable RCNN. To transfer general knowledge learned from data-abundant base classes to data-scarce novel classes, we propose a weight transfer strategy to promote model transferability and an attention-based feature enhancement mechanism to learn more robust object proposal feature representations. Further, we ensure strong discrimination by optimizing the contrastive objectives of feature maps via a supervised spatial contrastive loss. Meanwhile, we introduce an angle-guided additive margin classifier to augment instance-level inter-class difference and intra-class compactness, which is beneficial for improving the discriminative power of the few-shot classification head under a few supervisions. Our proposed framework outperforms the current works in various settings of PASCAL VOC and MSCOCO datasets; this demonstrates the effectiveness and generalization ability.

Object detection models based on feature pyramid networks have made significant progress in general object detection. However, small object detection is still a challenge for the existing models. In this paper, we think that two factors in the existing feature pyramid networks inhibit the performance of small object detection. The first one is that the different feature domains of shallow and deep layer features inhibit the model performance. The second one is that the accumulation of upper layer features leads to feature aliasing effect on the lower layer features, which interferes with the representations of small object features. Therefore, we propose Unified and Enhanced Feature Pyramid Networks (UEFPN) to improve the APs and ARs of small object detection. It has the following three characteristics: (1) Using the deep features of high-resolution image and original image to form the multi-scale features of unified domain. (2) In multi-scale features fusion, we learn the importance of upper layer features with the Channel Attention Fusion module (CAF) , to optimize feature aliasing effect and enhance the context information of shallow layer features. (3) UEFPN can be quickly applied to different models. The results of many experiments show that the models with UEFPN achieve significant performance improvement in small object detection compared with the baseline models.

In practical applications, object detection models often encounter new classes of interest after initial training. Traditional approaches require retraining from scratch using both existing and new class data, which becomes infeasible when original training images are inaccessible. Fine-tuning the model solely on new data typically leads to catastrophic forgetting - a severe performance degradation on previously learned classes. To address this challenge, we propose a novel Spatial Location-Aware Feature Distillation (SLFD) framework for Incremental Object Detection (IOD). Our method mitigates forgetting by distilling critical spatial features from the old model to the new one through two key mechanisms: (1) Spatial Transformation Feature Distillation (STFD), which identifies and transfers region-specific features from the old model, ensuring the incremental model retains spatial awareness of object locations; and (2) Spatial Attention Feature Distillation (SAFD), which preserves pixel-level knowledge by emphasizing salient spatial features. These mechanisms collectively enable the incremental model to maintain focus on object-related spatial features while minimizing forgetting. Extensive experiments on benchmark datasets under both single-step and multi-step incremental learning scenarios demonstrate that SLFD outperforms state-of-the-art methods, particularly in mitigating catastrophic forgetting in multi-step settings. Furthermore, evaluations in an autonomous driving scenario confirm SLFD’s practical effectiveness for real-world applications, showing robust detection of new objects while preserving prior knowledge.