![人工智能工具包（AI Toolkit）]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit.png/_jcr_content/renditions/original.image_file.100.100.file/ai-toolkit.png","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit.png/_jcr_content/renditions/original.image_file.360.360.file/ai-toolkit.png","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit.png/_jcr_content/renditions/original.image_file.768.768.file/ai-toolkit.png","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit.png/_jcr_content/renditions/original.image_file.1024.1024.file/ai-toolkit.png","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit.png/_jcr_content/renditions/original./ai-toolkit.png","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit.png/_jcr_content/renditions/original./ai-toolkit.png","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit.png/_jcr_content/renditions/original./ai-toolkit.png"} "人工智能工具包（AI Toolkit）")
人工智能工具包（AI Toolkit） 通过机器学习进行生命科学图像分析
在生命科学应用中，机器学习可以极大地提高图像分析的效率，降低人为错误的风险。该工具包包括图像去噪、图像分割和对象分类解决方案。
![通过机器学习改进分析工作流的每一个步骤]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit-denoising-segmentation-classification.png/_jcr_content/renditions/original.image_file.100.100.file/ai-toolkit-denoising-segmentation-classification.png","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit-denoising-segmentation-classification.png/_jcr_content/renditions/original.image_file.360.360.file/ai-toolkit-denoising-segmentation-classification.png","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit-denoising-segmentation-classification.png/_jcr_content/renditions/original.image_file.768.768.file/ai-toolkit-denoising-segmentation-classification.png","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit-denoising-segmentation-classification.png/_jcr_content/renditions/original.image_file.1024.1024.file/ai-toolkit-denoising-segmentation-classification.png","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit-denoising-segmentation-classification.png/_jcr_content/renditions/original./ai-toolkit-denoising-segmentation-classification.png","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit-denoising-segmentation-classification.png/_jcr_content/renditions/original./ai-toolkit-denoising-segmentation-classification.png","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-toolkit-denoising-segmentation-classification.png/_jcr_content/renditions/original./ai-toolkit-denoising-segmentation-classification.png"} "通过机器学习改进分析工作流的每一个步骤")
通过机器学习改进分析工作流的每一个步骤
从生图优化到对象识别和分类，各种功能应有尽有
图像分析是一个多步骤过程，包括生图处理、通过分割进行感兴趣结构识别，然后基于其特性对这些结构进行分类。ZEN AI Toolkit由ZEN Intellesis驱动，可为整个工作流的每一步提供工具。
- Intellesis去噪可以优化生图。
- Intellesis图像分割可以识别一个或多个对象分类
- Intellesis分类可以将这些对象分到适当的子组
![在U2OS+ LLC2细胞培养过程中对细胞表型的分类。一些细胞为手动分类]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification-training.jpg/_jcr_content/renditions/original.image_file.100.100.file/ai-classification-training.jpg","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification-training.jpg/_jcr_content/renditions/original.image_file.360.360.file/ai-classification-training.jpg","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification-training.jpg/_jcr_content/renditions/original.image_file.768.768.file/ai-classification-training.jpg","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification-training.jpg/_jcr_content/renditions/original./ai-classification-training.jpg","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification-training.jpg/_jcr_content/renditions/original./ai-classification-training.jpg","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification-training.jpg/_jcr_content/renditions/original./ai-classification-training.jpg","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification-training.jpg/_jcr_content/renditions/original./ai-classification-training.jpg"} "在U2OS+ LLC2细胞培养过程中对细胞表型的分类。一些细胞为手动分类")
![Intellesis分类接管了整个数据集的分类]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification.jpg/_jcr_content/renditions/original.image_file.100.100.file/ai-classification.jpg","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification.jpg/_jcr_content/renditions/original.image_file.360.360.file/ai-classification.jpg","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification.jpg/_jcr_content/renditions/original.image_file.768.768.file/ai-classification.jpg","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification.jpg/_jcr_content/renditions/original./ai-classification.jpg","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification.jpg/_jcr_content/renditions/original./ai-classification.jpg","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification.jpg/_jcr_content/renditions/original./ai-classification.jpg","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-classification.jpg/_jcr_content/renditions/original./ai-classification.jpg"} "Intellesis分类接管了整个数据集的分类")
深度学习简化自动图像分析
无需手动编程，新用户也可很快上手
传统的处理算法要求用户先编程，然后再对工具和参数进行微调，才能获得出色分析结果。这要求用户对图像处理算法和编程知识都有深入的了解，并进行长时间的试错实验，才能为特定实验找到合适的工具和参数。这一整套复杂的技术和时间要求通常让分析算法对许多研究人员来说难以企及。而借助机器学习，用户可以将理想结果示教于算法，算法即可找到解决问题的合适工具。整个过程非常快速、直观并且易于使用，即使是新用户也可掌握。
![标记图像以训练分割模型]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation-training.jpg/_jcr_content/renditions/original.image_file.100.100.file/ai-segmentation-training.jpg","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation-training.jpg/_jcr_content/renditions/original.image_file.360.360.file/ai-segmentation-training.jpg","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation-training.jpg/_jcr_content/renditions/original.image_file.768.768.file/ai-segmentation-training.jpg","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation-training.jpg/_jcr_content/renditions/original./ai-segmentation-training.jpg","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation-training.jpg/_jcr_content/renditions/original./ai-segmentation-training.jpg","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation-training.jpg/_jcr_content/renditions/original./ai-segmentation-training.jpg","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation-training.jpg/_jcr_content/renditions/original./ai-segmentation-training.jpg"} "标记图像以训练分割模型")
![基于该模型的自动图像分割]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation.jpg/_jcr_content/renditions/original.image_file.100.100.file/ai-segmentation.jpg","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation.jpg/_jcr_content/renditions/original.image_file.360.360.file/ai-segmentation.jpg","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation.jpg/_jcr_content/renditions/original.image_file.768.768.file/ai-segmentation.jpg","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation.jpg/_jcr_content/renditions/original./ai-segmentation.jpg","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation.jpg/_jcr_content/renditions/original./ai-segmentation.jpg","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation.jpg/_jcr_content/renditions/original./ai-segmentation.jpg","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/ai-segmentation.jpg/_jcr_content/renditions/original./ai-segmentation.jpg"} "基于该模型的自动图像分割")
全面的、不偏不倚的视图可提供更好的表现
数字图像可以提供高于单个像素的清晰度
显微镜采集的数字图像是一组像素值。传统算法一般会单独评估每个像素。但是，如果要更好地理解数字图像提供的数据，就不仅需要考虑每个像素本身，还要考虑其与其他像素的周边关系，这包括边缘、梯度、纹理和形状以及背景和阴影等图像伪影。机器学习能够在一定程度上避免人为造成的偏差和错误，以一种近乎不偏不倚的方式处理大量此类数据，从而提供更为优质可靠的结果。
蔡司Intellesis图像分割的技术详情
用于提升性能的独特方法
![深度学习与机器学习相结合：将VGG19神经网络馈入随机森林分类器]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/vgg19-deep-learning-random-forest-machine-learning.png/_jcr_content/renditions/original.image_file.100.100.file/vgg19-deep-learning-random-forest-machine-learning.png","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/vgg19-deep-learning-random-forest-machine-learning.png/_jcr_content/renditions/original.image_file.360.360.file/vgg19-deep-learning-random-forest-machine-learning.png","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/vgg19-deep-learning-random-forest-machine-learning.png/_jcr_content/renditions/original.image_file.768.768.file/vgg19-deep-learning-random-forest-machine-learning.png","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/vgg19-deep-learning-random-forest-machine-learning.png/_jcr_content/renditions/original.image_file.1024.1024.file/vgg19-deep-learning-random-forest-machine-learning.png","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/vgg19-deep-learning-random-forest-machine-learning.png/_jcr_content/renditions/original./vgg19-deep-learning-random-forest-machine-learning.png","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/vgg19-deep-learning-random-forest-machine-learning.png/_jcr_content/renditions/original./vgg19-deep-learning-random-forest-machine-learning.png","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/vgg19-deep-learning-random-forest-machine-learning.png/_jcr_content/renditions/original./vgg19-deep-learning-random-forest-machine-learning.png"} "深度学习与机器学习相结合：将VGG19神经网络馈入随机森林分类器")
深度学习与机器学习相结合：将VGG19神经网络馈入随机森林分类器
深度学习与机器学习相结合：将VGG19神经网络馈入随机森林分类器
深度学习与机器学习相结合：将VGG19神经网络馈入随机森林分类器
结合机器学习和深度学习的优势
随机森林分类器：多数票的力量
蔡司Intellesis图像分割采用名为“随机森林”的机器学习像素分类算法。该算法基于可依据大量像素特征对像素进行分类的决策树。随机森林使用许多这样的“树”，并根据多数票对像素进行分类。这样可以使分割非常可靠，所需的训练数据极少，并且训练过程比其他算法快许多。
将VGG19神经网络馈入随机森林分类器
标准随机森林分类器依赖的只有标准图像过滤器，因此可能缺少情境感知能力（“像素的周围要素”）。通过将图像输入VGG19神经网络并将特征图提供给随机森林分类器，可以将随机森林的速度与出色的深度神经网络图像识别功能相结合。如此即可获得更好的情境感知能力，从而得到更优质的分割结果。
|
|
|
|
|
|---|---|---|---|
|
开放式结构 |
ZEN工作流集成 |
性能 |
可靠的算法 |
|
|
|
|
![人工智能图像分析简介]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/intellesis-video-preview.jpg/_jcr_content/renditions/original.image_file.100.56.file/intellesis-video-preview.jpg","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/intellesis-video-preview.jpg/_jcr_content/renditions/original.image_file.360.203.file/intellesis-video-preview.jpg","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/intellesis-video-preview.jpg/_jcr_content/renditions/original.image_file.768.432.file/intellesis-video-preview.jpg","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/intellesis-video-preview.jpg/_jcr_content/renditions/original.image_file.1024.576.file/intellesis-video-preview.jpg","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/intellesis-video-preview.jpg/_jcr_content/renditions/original./intellesis-video-preview.jpg","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/intellesis-video-preview.jpg/_jcr_content/renditions/original./intellesis-video-preview.jpg","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/ai-toolkit/intellesis-video-preview.jpg/_jcr_content/renditions/original./intellesis-video-preview.jpg"} "人工智能图像分析简介")
人工智能图像分析简介
Intellesis机器学习
下载
-
ZEISS ZEN AI Toolkit
Segmentation and Classification by Machine Learning
文件大小: 1 MB
![ZEISS ZEN AI Toolkit的预览图像]({"max":"https://asset-downloads.zeiss.com/catalogs/preview/mic/42c3af22-64b8-4225-81fe-0160c4346e9a/EN_product-flyer_ZEN_AI-Toolkit.pdf.jpg"} "下载ZEISS ZEN AI Toolkit")
![3D工具包]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/3d-toolkit/3d-toolkit.png/_jcr_content/renditions/original.image_file.100.100.file/3d-toolkit.png","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/3d-toolkit/3d-toolkit.png/_jcr_content/renditions/original.image_file.360.360.file/3d-toolkit.png","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/3d-toolkit/3d-toolkit.png/_jcr_content/renditions/original.image_file.768.768.file/3d-toolkit.png","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/3d-toolkit/3d-toolkit.png/_jcr_content/renditions/original.image_file.1024.1024.file/3d-toolkit.png","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/3d-toolkit/3d-toolkit.png/_jcr_content/renditions/original./3d-toolkit.png","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/3d-toolkit/3d-toolkit.png/_jcr_content/renditions/original./3d-toolkit.png","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/3d-toolkit/3d-toolkit.png/_jcr_content/renditions/original./3d-toolkit.png"} "3D工具包")
![去卷积工具包]({"xsmall":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/deconvolution-toolkit/deconvolution-toolkit.png/_jcr_content/renditions/original.image_file.100.100.file/deconvolution-toolkit.png","small":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/deconvolution-toolkit/deconvolution-toolkit.png/_jcr_content/renditions/original.image_file.360.360.file/deconvolution-toolkit.png","medium":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/deconvolution-toolkit/deconvolution-toolkit.png/_jcr_content/renditions/original.image_file.768.768.file/deconvolution-toolkit.png","large":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/deconvolution-toolkit/deconvolution-toolkit.png/_jcr_content/renditions/original.image_file.1024.1024.file/deconvolution-toolkit.png","xlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/deconvolution-toolkit/deconvolution-toolkit.png/_jcr_content/renditions/original./deconvolution-toolkit.png","xxlarge":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/deconvolution-toolkit/deconvolution-toolkit.png/_jcr_content/renditions/original./deconvolution-toolkit.png","max":"https://www.zeiss.com/content/dam/rms/reference-master/products/software/zen/zen-desk/zen-software-packages/deconvolution-toolkit/deconvolution-toolkit.png/_jcr_content/renditions/original./deconvolution-toolkit.png"} "去卷积工具包")