JujubeNet: A high-precision lightweight jujube surface defect classification network with an attention mechanism

Abstract

Surface Defect Detection (SDD) is a significant research content in Industry 4.0 field. In the real complex industrial environment, SDD is often faced with many challenges, such as small difference between defect imaging and background, low contrast, large variation of defect scale and diverse types, and large amount of noise in defect images. Jujubes are naturally growing plants, and the appearance of the same type of surface defect can vary greatly, so it is more difficult than industrial products produced according to the prescribed process. In this paper, a ConvNeXt-based high-precision lightweight classification network JujubeNet is presented to address the practical needs of Jujube Surface Defect (JSD) classification. In the proposed method, a Multi-branching module using Depthwise separable Convolution (MDC) is designed to extract more feature information through multi-branching and substantially reduces the number of parameters in the model by using depthwise separable convolutions. What’s more, in our proposed method, the Convolutional Block Attention Module (CBAM) is introduced to make the model concentrate on different classes of JSD features. The proposed JujubeNet is compared with other mainstream networks in the actual production environment. The experimental results show that the proposed JujubeNet can achieve 99.1% classification accuracy, which is significantly better than the current mainstream classification models. The FLOPS and parameters are only 30.7% and 30.6% of ConvNeXt-Tiny respectively, indicating that the model can quickly and effectively classify JSD and is of great practical value.

MpoxNet: dual-branch deep residual squeeze and excitation monkeypox classification network with attention mechanism

“…In terms of improving accuracy, increasing the cardinality of the network is more effective than increasing depth or width. This viewpoint was initially proposed in ResNeXt ( Xie et al., 2017 ; Jiang et al., 2023 ). Inspired by this, this paper introduces a Dual-Branch Depthwise Separable Convolution Residual Squeeze and Excitation (D 2 RSE) module in ConvNext.…”

DDRA-Net: Dual-Channel Deep Residual Attention UPerNet for Breast Lesions Segmentation in Ultrasound Images

“…In the quest for improved accuracy, the strategy of augmenting cardinality surpasses the effectiveness of expanding network depth or width. This conceptual innovation was originaly introduced in the influential ResNeXt framework [24], [25]. Drawing inspiration from this insight, the current study introduces a dual-channel structure in the UPerNet block.…”

“…This innovative design efficiently extracts features with limited parameters, intricately fuses multiscale semantic features, and successfully restores rich details through a unique attention mechanism, significantly improving image segmentation performance. Lin et al [21] introduced a deep dual attention network (D2ANet) for COVID-19 diagnosis using chest CT images, skillfully integrating dual attention modules(DAM) and multi-scale feature extractors to automatically detect lesion areas and extract [25]. While dilated convolutions and atention mechanisms can enhance the performance of segmentation netwo-rks, they still have some limitations.…”

“…The paper delves into the embedding strategy of CBAM in the Efficient Last Stage of the backbone network MobileNet V3 and explores three different structures [25], [26], as illustr ated in Fig. 5: (a…”

A high-precision jujube disease spot detection based on SSD during the sorting process

“…By analyzing the Jujube black spot disease dataset, we observed that small lesions accounted for 14.2% of the black spot disease samples, medium-sized lesions accounted for 42.8%, and large-sized lesions accounted for 20.9% of the dataset. Accurately detecting small lesions presents a significant challenge in identifying and diagnosing jujube black spot disease [31][32][33]. First, small lesions often manifest as tiny spots or patches on the jujube surface, exhibiting similar colors and shapes to the surrounding normal tissues, making visual observation and differentiation difficult without magnification.…”