农业病虫害
中国农业大学_叶菜病虫害图像识别挑战赛
中国农业大学_叶菜病虫害图像识别挑战赛
MatterPort Mask R-CNN for Here We Grow!
"Complete Guide to Creating COCO Datasets"
MNIST
这是一个包含著名MNIST数据集的仓库。卷积神经网络在Keras中被应用，以达到99%的准确率。
Nearest Neighbors Distance Ratio Open-Set Classifier
Feature vectors used in "Nearest neighbors distance ratio open-set classifier" paper to appear in Springer Machine Learning journal. In the 15-Scenes (15scenes.dat) dataset, with 15 classes, the 4,485 images were represented by a bag-of-visual-word vector created with soft assignment and max pooling, based on a codebook of 1,000 Scale Invariant Feature Transform (SIFT) codewords. The 26 classes of the letter (letter.dat) dataset represent the letters of the English alphabet (black-and-white rectangular pixel displays). The 20,000 samples contain 16 attributes. The Auslan (auslan.dat) dataset contains 95 classes of Australian Sign Language (Auslan) signs collected from a volunteer native Auslan signer. Data was acquired using two Fifth Dimension Technologies (5DT) gloves hardware and two Ascension Flock-of-Birds magnetic position trackers. There are 146,949 samples represented with 22 features (x, y, z positions, bend measures, etc). The Caltech-256 (caltech256.dat) dataset comprises 256 object classes. The feature vectors consider a bag-of-visual-words characterization approach and contain 1,000 features, acquired with dense sampling, SIFT descriptor for the points of interest, hard assignment, and average pooling. In total, there are 29,780 samples. The ALOI (aloi.dat) dataset has 1,000 classes and 108 samples for each class (108,000 in total). The features were extracted with the Border/Interior (BIC) descriptor and contain 128 dimensions. The ukbench (ukbench.dat) dataset comprises 2,550 classes of four images each. In our work, the images were represented with BIC descriptor (128 dimensions).
Synthetic Turfgrass Dataset
Synthetic Turfgrass Dataset created in blender, top view of a golf course surface. Image Type Folder name Count RGB Training Images Images 6510 Mono 8bit Annotation images: - Class grass, Value :30 - Class divot anomaly, Value: 10 ImageLables 6510
ImageNet-11
ICLR2022 Explainable AI: Object Recognition With Help From Background
WhichDog: A crowdsourced dataset including candidate set-based labelling
A dataset with crowdsourced labels for aggregation and supervised classification. It contains 400 images of dogs from the Stanford Dogs dataset (http://vision.stanford.edu/aditya86/ImageNetDogs/). Images of dogs that belong to 32 different breeds (classes) are included. Annotators were asked to provide two types of labelling: full labelling (each labeler is allowed to provide a single label for each image) and candidate labelling (each labeler is allowed to provide a set of candidate labels for each image). It includes a total of 61227 annotations (30628 full and 30599 candidate) obtained from a set of 1028 different labelers. The labels were collected through the online crowdsourcing platform Amazon mTurk thanks to funds provided by the Basque Government through the Elkartek program (KK-2018/00071). The assignments were designed as sequences of 64 images that were given to the annotators. Each image in the sequence was provided together with a specific subset of possible labels (with the number of options ranging from 4 to 32), and a instruction for the annotator to perform a specific type of labelling (full or candidate). Each labeler performed at least one assignment. Not all the labelers completed the 64 annotations in their assignments. The file 'whichdog.zip' contains a folder ('images') with the 400 images of dogs, a text file ('breed_names.txt') that indicates the names of the different breeds and their assigned label (a number in the interval from 0 to 31) and a CSV file ('whichdog_all_annots.csv') that contains the information about the annotations. Each row of the CSV file represents a single annotation, and each column shows: - image_id: ID number of the image. - is_candidate: indicates whether the requested labelling is full (0) or candidate (1). - labeler_id: ID number of the labeler. - time: time employed by the labeler to perform the annotation. - answer: label or set of labels provided by the labeler as annotation. - options: subset of possible labels shown to the labeler. - assignment_id: ID number of the assignment - sequence_point: number that indicates the point of the sequence of images of the assignment in which the annotation was provided. - class: ground truth label of the image.
Contour Identification method for human-robot communication
The subject of research was task-oriented communication between humans and collaborative robots in industrial scenario Human-robot interface must be supported by image recognition system in order to identify objects referred to by operator A novel method for contour classification based on flexible editable contour templates (FECT) was developed Practical result was a tool facilitating customization of image recognition systems Verification of the method was done through the laboratory implementation
MASKRCNN_TF_V2
MASK RCNN upgraded to Tensorflow V2
Annotated Chemical Apparatus Image Dataset
This dataset consists of 5078 annotated images of chemical apparatuses. These images were obtained by extracting diverse frames from videos of chemical experiments recorded using a smartphone. Annotations cover six experimental apparatuses and the experimenter's hands. The dataset is divided into three se ts: training, validation, and test, each with accompanying annotation texts in YOLO (You Only Look Once) format.
Cat vs Rabbit Mobile Net Predictor | Acc : 100%
Cat vs Rabbit Classification problem solved with 100% accuracy of Test Set.
Camera Photos vs Ai generated Photos Classifier
Camera clicked photo vs ai generated photos classifier
Seadragon survey data
Data represents seadragon surveys over 8 years. All seadragons individuals are uniquely identified using image spot pattern recognition program I3S.
EfficientNetB0-7_NoTop_imagenet
EfficientNetB 0 to 7_ NoTop with imagenet pretrained weights
Major and minor chords
Standardized images of sheet music representing major and minor chords.
DogVSTortoise Dataset for Kaggle
这是一个专为Kaggle竞赛定制的数据集。我收集并标注了20,000张图片，其中一半是狗，一半是乌龟。该数据集对所有人免费开放。由于数据集体积过大，无法上传至GitHub，因此我将其放置在百度网盘上。
trained_cassava_weights
trained EfficientNetB5 weights for cassava
Leaf Diseases
Collection of various leaf images with different types of diseases
papaya
**Data Description:** The dataset contains over 500 images of Carica papaya, categorized as either "good" or "bad". These images were captured using a Realme 11x Next Era smartphone camera, ensuring consistent image quality and resolution. Each image features a single Carica papaya fruit against a white background, with the data collection process conducted under natural daylight conditions for optimal illumination. **Key Features:** 1. **Image Variation:** The dataset encompasses a diverse range of Carica papaya fruits, capturing variations in size, shape, color, ripeness, and overall condition. This variability is crucial for training robust classification models capable of accurately distinguishing between good and bad fruits. 2. **Annotation:** Each image in the dataset is annotated to indicate whether the Carica papaya fruit is categorized as "good" or "bad". These annotations provide ground truth labels for supervised learning algorithms, facilitating the development of accurate classification models. 3. **Consistent Background:** To ensure uniformity and minimize distractions, all images feature a white background. This consistent background simplifies preprocessing and enables the focus to remain solely on the visual attributes of the Carica papaya fruits. 4. **Daylight Conditions:** The data collection process was conducted under natural daylight conditions to ensure consistent illumination across all images. Natural light enhances the visibility of fruit features and minimizes lighting variations, contributing to the authenticity and quality of the dataset. 5. **High-Quality Images:** Images captured with the Realme 11x Next Era smartphone camera exhibit high resolution and clarity, enabling detailed analysis of fruit characteristics. The quality of the images facilitates precise feature extraction, essential for accurate classification. 6. **Large Dataset Size:** With over 500 images, the dataset provides a significant volume of data for model training and validation. A larger dataset enhances model generalization and reduces the risk of overfitting, leading to improved classification performance on unseen data. **Potential Applications:** 1. **Automated Fruit Classification:** The dataset can be used to develop machine learning models capable of automatically classifying Carica papaya fruits as "good" or "bad" based on their visual characteristics. Such models can assist in quality assessment and sorting processes in agriculture or food industries. 2. **Fruit Quality Control:** By analyzing fruit condition through automated classification, farmers and producers can monitor the quality of Carica papaya fruits and detect potential defects or ripeness issues early. Timely interventions can then be implemented to optimize fruit yield and quality.
Shark species
collection of shark images classified by 14 different species
中国农业大学_叶菜病虫害图像识别挑战赛
"Complete Guide to Creating COCO Datasets"
这是一个包含著名MNIST数据集的仓库。卷积神经网络在Keras中被应用，以达到99%的准确率。
Feature vectors used in "Nearest neighbors distance ratio open-set classifier" paper to appear in Springer Machine Learning journal. In the 15-Scenes (15scenes.dat) dataset, with 15 classes, the 4,485 images were represented by a bag-of-visual-word vector created with soft assignment and max pooling, based on a codebook of 1,000 Scale Invariant Feature Transform (SIFT) codewords. The 26 classes of the letter (letter.dat) dataset represent the letters of the English alphabet (black-and-white rectangular pixel displays). The 20,000 samples contain 16 attributes. The Auslan (auslan.dat) dataset contains 95 classes of Australian Sign Language (Auslan) signs collected from a volunteer native Auslan signer. Data was acquired using two Fifth Dimension Technologies (5DT) gloves hardware and two Ascension Flock-of-Birds magnetic position trackers. There are 146,949 samples represented with 22 features (x, y, z positions, bend measures, etc). The Caltech-256 (caltech256.dat) dataset comprises 256 object classes. The feature vectors consider a bag-of-visual-words characterization approach and contain 1,000 features, acquired with dense sampling, SIFT descriptor for the points of interest, hard assignment, and average pooling. In total, there are 29,780 samples. The ALOI (aloi.dat) dataset has 1,000 classes and 108 samples for each class (108,000 in total). The features were extracted with the Border/Interior (BIC) descriptor and contain 128 dimensions. The ukbench (ukbench.dat) dataset comprises 2,550 classes of four images each. In our work, the images were represented with BIC descriptor (128 dimensions).
Synthetic Turfgrass Dataset created in blender, top view of a golf course surface. Image Type Folder name Count RGB Training Images Images 6510 Mono 8bit Annotation images: - Class grass, Value :30 - Class divot anomaly, Value: 10 ImageLables 6510
ICLR2022 Explainable AI: Object Recognition With Help From Background
A dataset with crowdsourced labels for aggregation and supervised classification. It contains 400 images of dogs from the Stanford Dogs dataset (http://vision.stanford.edu/aditya86/ImageNetDogs/). Images of dogs that belong to 32 different breeds (classes) are included. Annotators were asked to provide two types of labelling: full labelling (each labeler is allowed to provide a single label for each image) and candidate labelling (each labeler is allowed to provide a set of candidate labels for each image). It includes a total of 61227 annotations (30628 full and 30599 candidate) obtained from a set of 1028 different labelers. The labels were collected through the online crowdsourcing platform Amazon mTurk thanks to funds provided by the Basque Government through the Elkartek program (KK-2018/00071). The assignments were designed as sequences of 64 images that were given to the annotators. Each image in the sequence was provided together with a specific subset of possible labels (with the number of options ranging from 4 to 32), and a instruction for the annotator to perform a specific type of labelling (full or candidate). Each labeler performed at least one assignment. Not all the labelers completed the 64 annotations in their assignments. The file 'whichdog.zip' contains a folder ('images') with the 400 images of dogs, a text file ('breed_names.txt') that indicates the names of the different breeds and their assigned label (a number in the interval from 0 to 31) and a CSV file ('whichdog_all_annots.csv') that contains the information about the annotations. Each row of the CSV file represents a single annotation, and each column shows: - image_id: ID number of the image. - is_candidate: indicates whether the requested labelling is full (0) or candidate (1). - labeler_id: ID number of the labeler. - time: time employed by the labeler to perform the annotation. - answer: label or set of labels provided by the labeler as annotation. - options: subset of possible labels shown to the labeler. - assignment_id: ID number of the assignment - sequence_point: number that indicates the point of the sequence of images of the assignment in which the annotation was provided. - class: ground truth label of the image.
The subject of research was task-oriented communication between humans and collaborative robots in industrial scenario Human-robot interface must be supported by image recognition system in order to identify objects referred to by operator A novel method for contour classification based on flexible editable contour templates (FECT) was developed Practical result was a tool facilitating customization of image recognition systems Verification of the method was done through the laboratory implementation
MASK RCNN upgraded to Tensorflow V2
This dataset consists of 5078 annotated images of chemical apparatuses. These images were obtained by extracting diverse frames from videos of chemical experiments recorded using a smartphone. Annotations cover six experimental apparatuses and the experimenter's hands. The dataset is divided into three se ts: training, validation, and test, each with accompanying annotation texts in YOLO (You Only Look Once) format.
Cat vs Rabbit Classification problem solved with 100% accuracy of Test Set.
Camera clicked photo vs ai generated photos classifier
Data represents seadragon surveys over 8 years. All seadragons individuals are uniquely identified using image spot pattern recognition program I3S.
EfficientNetB 0 to 7_ NoTop with imagenet pretrained weights
Standardized images of sheet music representing major and minor chords.
这是一个专为Kaggle竞赛定制的数据集。我收集并标注了20,000张图片，其中一半是狗，一半是乌龟。该数据集对所有人免费开放。由于数据集体积过大，无法上传至GitHub，因此我将其放置在百度网盘上。
trained EfficientNetB5 weights for cassava
Collection of various leaf images with different types of diseases
**Data Description:** The dataset contains over 500 images of Carica papaya, categorized as either "good" or "bad". These images were captured using a Realme 11x Next Era smartphone camera, ensuring consistent image quality and resolution. Each image features a single Carica papaya fruit against a white background, with the data collection process conducted under natural daylight conditions for optimal illumination. **Key Features:** 1. **Image Variation:** The dataset encompasses a diverse range of Carica papaya fruits, capturing variations in size, shape, color, ripeness, and overall condition. This variability is crucial for training robust classification models capable of accurately distinguishing between good and bad fruits. 2. **Annotation:** Each image in the dataset is annotated to indicate whether the Carica papaya fruit is categorized as "good" or "bad". These annotations provide ground truth labels for supervised learning algorithms, facilitating the development of accurate classification models. 3. **Consistent Background:** To ensure uniformity and minimize distractions, all images feature a white background. This consistent background simplifies preprocessing and enables the focus to remain solely on the visual attributes of the Carica papaya fruits. 4. **Daylight Conditions:** The data collection process was conducted under natural daylight conditions to ensure consistent illumination across all images. Natural light enhances the visibility of fruit features and minimizes lighting variations, contributing to the authenticity and quality of the dataset. 5. **High-Quality Images:** Images captured with the Realme 11x Next Era smartphone camera exhibit high resolution and clarity, enabling detailed analysis of fruit characteristics. The quality of the images facilitates precise feature extraction, essential for accurate classification. 6. **Large Dataset Size:** With over 500 images, the dataset provides a significant volume of data for model training and validation. A larger dataset enhances model generalization and reduces the risk of overfitting, leading to improved classification performance on unseen data. **Potential Applications:** 1. **Automated Fruit Classification:** The dataset can be used to develop machine learning models capable of automatically classifying Carica papaya fruits as "good" or "bad" based on their visual characteristics. Such models can assist in quality assessment and sorting processes in agriculture or food industries. 2. **Fruit Quality Control:** By analyzing fruit condition through automated classification, farmers and producers can monitor the quality of Carica papaya fruits and detect potential defects or ripeness issues early. Timely interventions can then be implemented to optimize fruit yield and quality.
collection of shark images classified by 14 different species
查看更多数据集
相关知识
农业病虫害网络化远程诊断平台国家（AA)农业病虫害网络
智慧农业病虫害预警系统
农业病虫害专家诊断
智慧农业病虫害系统 虫情测报系统 农业病虫害防御系统
农业病虫害智能识别系统.docx
茶树病虫害农业防治
农业病虫害远程诊断平台
农业病虫害网络化远程诊断平台
《桃树病虫害》：生态平衡与农业
农业病虫害预测预报.docx
网址: 农业病虫害 https://m.huajiangbk.com/newsview480486.html
| 上一篇: 【招标】乌鲁木齐市蓝天森林花苑智 |
下一篇: 智能养殖监控系统，自动化管理养殖 |
