Online Detection System for Wheat Machine Harvesting Impurity Rate Based on DeepLabV3+

Abstract

Wheat, one of the most important food crops in the world, is usually harvested mechanically by combine harvesters. The impurity rate is one of the most important indicators of the quality of wheat obtained by mechanized harvesting. To realize the online detection of the impurity rate in the mechanized harvesting process of wheat, a vision system based on the DeepLabV3+ model of deep learning for identifying and segmenting wheat grains and impurities was designed in this study. The DeepLabV3+ model construction considered the four backbones of MobileNetV2, Xception-65, ResNet-50, and ResNet-101 for training. The optimal DeepLabV3+ model was determined through the accuracy rate, comprehensive evaluation index, and average intersection ratio. On this basis, an online detection method of measuring the wheat impurity rate in mechanized harvesting based on image information was constructed. The model realized the online detection of the wheat impurity rate. The test results showed that ResNet-50 had the best recognition and segmentation performance; the accuracy rate of grain identification was 86.86%; the comprehensive evaluation index was 83.63%; the intersection ratio was 0.7186; the accuracy rate of impurity identification was 89.91%; the comprehensive evaluation index was 87.18%; the intersection ratio was 0.7717; and the average intersection ratio was 0.7457. In terms of speed, ResNet-50 had a fast segmentation speed of 256 ms per image. Therefore, in this study, ResNet-50 was selected as the backbone network for DeepLabV3+ to carry out the identification and segmentation of mechanically harvested wheat grains and impurity components. Based on the manual inspection results, the maximum absolute error of the device impurity rate detection in the bench test was 0.2%, and the largest relative error was 17.34%; the maximum absolute error of the device impurity rate detection in the field test was 0.06%; and the largest relative error was 13.78%. This study provides a real-time method for impurity rate measurement in wheat mechanized harvesting.

Detection of breakage and impurity ratios for raw sugarcane based on estimation model and MDSC-DeepLabv3+

“…Semantic segmentation, on the other hand, enables pixel-wise classification of an image and facilitates the precise determination of the number of pixels and their respective categories in a specific region. Mass-pixel fitting models can be established by combining the number of pixels and the actual mass of each category of object (Rong et al, 2020;Chen et al, 2022;Jin et al, 2022;Liu et al, 2022;Liu et al, 2023), thus supporting the quantitative analysis of the quality of the detected objects. In order to quantify the ratio of breakage and impurity in raw sugarcane, semantic segmentation technology was utilized to abstract the of raw sugarcane and impurities in this study.…”

“…In general, previous estimation models that are based on image pixels for assessing the mass of crops (such as wheat, corn, and soybean) often assume that the surface density (mass/pixel) of each crop category remains stable across different images (Chen et al, 2022;Jin et al, 2022;Liu et al, 2022). However, when it comes to broken cane and impurities, their surface density can vary significantly in different images.…”

“…To improve accuracy in wheat and impurity recognition, Shen et al (2019) constructed a dataset and trained a recognition model called WheNet based on Inception_v3, achieving a recall rate of 98% and an efficiency of 100ms per image. Chen et al (2022) designed a vision system based on DeepLabv3+ to identify seeds and impurities in wheat, obtaining mean pixel accuracy (mPA) values of 86.86% and 89.91% for grains and impurities, and mean intersection over union (mIoU) scores of 0.7186 and 0.7457, respectively. For the detection of impurities in the corn deep-bed drying process, employed a multiscale color recovery algorithm to enhance images and eliminate noise.…”

From Segmentation to Classification: A Deep Learning Scheme for Sintered Surface Images Processing

“…DeepLabV3+, as a mature semantic segmentation model, is widely used in various tasks [25][26][27][28][29]. However, it suffers from high complexity and computational resource consumption.…”

Stripe pooling and densely connected DeepLabv3 plus efficient semantic segmentation

“…The overall structure of the improved DeepLabv3 plus [8,9,10] model is shown in Figure 1 below. In the encoder section, MobileneV2 is used as the main part to accelerate prediction speed.…”