智能检测管制刀具模型X光图像训练数据

本数据集包含多角度、多场景下的管制刀具的X光安检图像，通过对图像的标注、抠图、增强、融合等处理，可作为优质样本训练生成管制刀具的智能检测模型，实现算法模型对管制刀具的精准识别。本数据集管制刀具的图例形态丰富、场景多样、更新及时。通过本数据集的深入应用，可有效提升管制刀具智能检验检测模型的检测精度、速度，提升模型性能。1、数据来源：应用X射线光源多维度、多角度、多场景下透射管制刀具采集并建立其原始的X光数据图例库。 2、数据深度处理：对采集到的原始X光图像预标注坐标位置和品项类别，并对管制刀具图像进行抠图处理。将抠出的管制刀具图像与多场景的图像分别进行几何变换、像素变换等增广处理。 3、检测模型生成算法规则：将处理后的管制刀具X光图像和场景图像通过密度统计（像素值代表实物密度值）依据区域匹配原则进行融合，融合区域掩模作为数据标签与融合后的图像作为深度学习样本数据。还可通过调整抠图区域在场景图像区域的位置，获得不同的平均密度差值，训练生成可精准定位、精准识别管制刀具的智能检测模型。区域匹配原则按照Mask*(α*ρ抠图图像+β*ρ场景图像)，融合后的图像处理公式按照Mask*(α*ρ抠图图像+β*ρ抠图图像)+（1-Mask）*ρ场景图像。（所述公式中：Mask为图像掩膜，图像目标区域值为1，目标区域外值为0，ρ为密度值，α、β指系数）检测模型可对多场景下的管制刀具物品精准识别，同时将目标物的位置及所在X光图像信息记录标出。进一步的还可根据目标物位置信息推算目标物尺寸信息。

This dataset contains X-ray security inspection images of prohibited knives collected under multi-dimensional, multi-angle and multi-scenario conditions. Through processing including image annotation, matting, data augmentation and fusion, it can be used as high-quality samples to train and develop intelligent detection models for prohibited knives, enabling algorithm models to achieve accurate identification of such knives. The dataset features diverse shapes and forms of prohibited knives, varied application scenarios, and timely updates. In-depth application of this dataset can effectively improve the detection accuracy, speed and overall performance of intelligent inspection and detection models for prohibited knives.

1. Data Source: Collect transmitted X-ray images of prohibited knives under multi-dimensional, multi-angle and multi-scenario conditions using X-ray sources, and establish an original X-ray image database of prohibited knives.
2. In-depth Data Processing: Pre-label the coordinate positions and category types of the collected original X-ray images, and perform matting processing on the images of prohibited knives. Then, conduct data augmentation operations including geometric transformation and pixel transformation on the matting-extracted prohibited knife images and multi-scenario images respectively.
3. Algorithm Rules for Generating Detection Models: Fuse the processed X-ray images of prohibited knives and scenario images based on the region matching principle via density statistics (pixel values represent the density values of physical objects). The fusion region mask serves as the data label, while the fused image serves as the deep learning sample data. Additionally, by adjusting the position of the matting-extracted region within the scenario image, different average density differences can be obtained to train and generate intelligent detection models that can accurately position and identify prohibited knives. The region matching principle follows the formula: Mask*(α*ρ_matting_image + β*ρ_matting_image). The image processing formula for the fused result follows: Mask*(α*ρ_matting_image + β*ρ_matting_image) + (1-Mask)*ρ_scenario_image. In the formulas mentioned above: Mask is the image mask, with a value of 1 in the target region of the image and 0 outside the target region; ρ represents the density value; α and β are coefficients. The detection model can accurately identify prohibited knives in various scenarios, and record and mark the position of the target object and the information of the X-ray image where it is located. Furthermore, the size information of the target object can be inferred based on its position information.

AI搜集汇总

数据集介绍

特点

该数据集是一个包含2620条X光安检图像的训练数据，用于智能检测管制刀具模型。图像经过多角度、多场景采集和深度处理，适用于提升模型检测精度和速度。