The global image tagging and annotation services market is experiencing robust growth, driven by the increasing adoption of artificial intelligence (AI) and machine learning (ML) across diverse sectors. The market, estimated at $2.5 billion in 2025, is projected to expand at a Compound Annual Growth Rate (CAGR) of 18% from 2025 to 2033, reaching an estimated $10 billion by 2033. This significant expansion is fueled by several key factors. The automotive industry leverages image tagging and annotation for autonomous vehicle development, requiring vast amounts of labeled data for training AI algorithms. Similarly, the retail and e-commerce sectors utilize these services for image search, product recognition, and improved customer experiences. The healthcare industry benefits from advancements in medical image analysis, while the government and security sectors employ image annotation for surveillance and security applications. The rising availability of high-quality data, coupled with the decreasing cost of annotation services, further accelerates market growth. However, challenges remain. Data privacy concerns and the need for high-accuracy annotation can pose significant hurdles. The demand for specialized skills in data annotation also contributes to a potential bottleneck in the market's growth trajectory. Overcoming these challenges requires a collaborative approach, involving technological advancements in automation and the development of robust data governance frameworks. The market segmentation, encompassing various annotation types (image classification, object recognition/detection, boundary recognition, segmentation) and application areas (automotive, retail, BFSI, government, healthcare, IT, transportation, etc.), presents diverse opportunities for market players. The competitive landscape includes a mix of established players and emerging firms, each offering specialized services and targeting specific market segments. North America currently holds the largest market share due to early adoption of AI and ML technologies, while Asia-Pacific is anticipated to witness rapid growth in the coming years.

Here are a few use cases for this project:

1. Educational Applications: This model can be utilized in interactive learning applications to help children understand different shapes, their orientations, and combinations in a fun and engaging way. The model can quickly identify and provide feedback on shapes drawn by children.

2. Quality Control in Manufacturing: The object detection model can be deployed on assembly lines or in manufacturing facilities to identify specific markers to ensure components are correctly oriented during assembly or produce. With different shapes representing different components or assembly stages, the model can reduce human error and speed up the production process.

3. Image-tagging Software: This model can be used in a software application that auto-tags or auto-categorizes pictures based on the markers present in them. This will improve search and retrieval of images in large databases.

4. Augmented Reality (AR) Apps: In AR games or applications, this model can identify markers to trigger certain actions or reactions, creating a more immersive and interactive AR experience. For example, once a specific shape marker is detected in the user's environment, a corresponding AR object can appear.

5. Robotics and Navigation: Robotic systems can use the model to understand visual cues for path setting and maneuvering in their environment. This is particularly helpful in settings where different shapes indicate different paths or actions to be taken by an autonomous system. Certain oriented markers can be used to guide robots on their route and motion.

Image Annotation Tool Market Outlook

The global image annotation tool market size is projected to grow from approximately $700 million in 2023 to an estimated $2.5 billion by 2032, exhibiting a remarkable compound annual growth rate (CAGR) of 15.2% over the forecast period. The surging demand for machine learning and artificial intelligence applications is driving this robust market expansion. Image annotation tools are crucial for training AI models to recognize and interpret images, a necessity across diverse industries.

One of the key growth factors fueling the image annotation tool market is the rapid adoption of AI and machine learning technologies across various sectors. Organizations in healthcare, automotive, retail, and many other industries are increasingly leveraging AI to enhance operational efficiency, improve customer experiences, and drive innovation. Accurate image annotation is essential for developing sophisticated AI models, thereby boosting the demand for these tools. Additionally, the proliferation of big data analytics and the growing necessity to manage large volumes of unstructured data have amplified the need for efficient image annotation solutions.

Another significant driver is the increasing use of autonomous systems and applications. In the automotive industry, for instance, the development of autonomous vehicles relies heavily on annotated images to train algorithms for object detection, lane discipline, and navigation. Similarly, in the healthcare sector, annotated medical images are indispensable for developing diagnostic tools and treatment planning systems powered by AI. This widespread application of image annotation tools in the development of autonomous systems is a critical factor propelling market growth.

The rise of e-commerce and the digital retail landscape has also spurred demand for image annotation tools. Retailers are using these tools to optimize visual search features, personalize shopping experiences, and enhance inventory management through automated recognition of products and categories. Furthermore, advancements in computer vision technology have expanded the capabilities of image annotation tools, making them more accurate and efficient, which in turn encourages their adoption across various industries.

Data Annotation Software plays a pivotal role in the image annotation tool market by providing the necessary infrastructure for labeling and categorizing images efficiently. These software solutions are designed to handle various annotation tasks, from simple bounding boxes to complex semantic segmentation, enabling organizations to generate high-quality training datasets for AI models. The continuous advancements in data annotation software, including the integration of machine learning algorithms for automated labeling, have significantly enhanced the accuracy and speed of the annotation process. As the demand for AI-driven applications grows, the reliance on robust data annotation software becomes increasingly critical, supporting the development of sophisticated models across industries.

Regionally, North America holds the largest share of the image annotation tool market, driven by significant investments in AI and machine learning technologies and the presence of leading technology companies. Europe follows, with strong growth supported by government initiatives promoting AI research and development. The Asia Pacific region presents substantial growth opportunities due to the rapid digital transformation in emerging economies and increasing investments in technology infrastructure. Latin America and the Middle East & Africa are also expected to witness steady growth, albeit at a slower pace, due to the gradual adoption of advanced technologies.

Component Analysis

The image annotation tool market by component is segmented into software and services. The software segment dominates the market, encompassing a variety of tools designed for different annotation tasks, from simple image labeling to complex polygonal, semantic, or instance segmentation. The continuous evolution of software platforms, integrating advanced features such as automated annotation and machine learning algorithms, has significantly enhanced the accuracy and efficiency of image annotations. Furthermore, the availability of open-source annotation tools has lowered the entry barrier, allowing more organizations to adopt these technologies.

Services associated with image ann

The global image capture and processing software market is experiencing robust growth, driven by the increasing adoption of digital technologies across diverse sectors. The market's expansion is fueled by several key factors, including the proliferation of smartphones with advanced camera capabilities, the rising demand for high-quality images in various applications (from social media to professional photography), and the increasing need for efficient image analysis tools in fields like medicine and security. The market is segmented by application (financial, medical, security monitoring, smart home, and others) and software type (image capture, recognition & analysis, editing, and comprehensive solutions). While established players like Adobe, Corel, and Microsoft dominate the market, innovative startups and specialized companies are also emerging, particularly in the areas of AI-powered image recognition and analysis. This competitive landscape fosters continuous improvement in software capabilities and affordability, further stimulating market growth. A projected CAGR of 15% (a reasonable estimate given the rapid technological advancements in this space) suggests a significant market expansion over the forecast period (2025-2033). The North American and European markets currently hold significant shares, owing to established technological infrastructure and high adoption rates. However, the Asia-Pacific region is expected to witness the fastest growth, driven by increasing smartphone penetration, expanding internet access, and rising demand for digital imaging solutions in developing economies like India and China. Challenges to market growth include data security concerns, particularly around sensitive images used in medical and financial applications, and the need for robust software that can efficiently handle increasingly large image datasets. Despite these constraints, the overall market outlook remains highly positive, with substantial growth potential fueled by ongoing technological innovations and broadening application across various industries. The integration of artificial intelligence and machine learning capabilities within image processing software is a key trend, enhancing functionalities such as automated image tagging, object recognition, and advanced editing features.

Image Recognition Market Outlook

As per our latest research, the global image recognition market size reached USD 41.2 billion in 2024, reflecting robust demand across multiple industries. The market is expected to grow at a compound annual growth rate (CAGR) of 18.7% from 2025 to 2033, reaching an estimated value of USD 203.7 billion by 2033. This growth is propelled by increasing adoption of artificial intelligence (AI) and machine learning technologies, rising security and surveillance requirements, and the proliferation of smart devices equipped with advanced cameras and sensors.

One of the primary growth drivers for the image recognition market is the rapid evolution and integration of AI and deep learning algorithms in image analysis. Industries such as healthcare, automotive, and retail are leveraging these advancements to automate complex processes, improve accuracy, and enhance customer experiences. For instance, in healthcare, image recognition is being utilized for diagnostic imaging, enabling faster and more precise detection of diseases. Similarly, the automotive sector is deploying image recognition for autonomous vehicles and advanced driver assistance systems (ADAS), which rely heavily on real-time image processing and object detection to ensure road safety. This technological leap has significantly expanded the application scope and commercial viability of image recognition solutions.

Another significant factor contributing to market expansion is the exponential growth of digital content and the need for efficient management and analysis of visual data. With the proliferation of smartphones, social media platforms, and connected devices, there is an unprecedented volume of images and videos being generated daily. Organizations are increasingly adopting image recognition tools to automate the sorting, tagging, and retrieval of this data, thereby streamlining operations and gaining actionable insights. In the retail and e-commerce sector, for example, image recognition powers visual search, personalized recommendations, and inventory management, which directly impact customer engagement and operational efficiency. The ongoing digital transformation across industries is expected to further accelerate the adoption of image recognition technologies globally.

The rising emphasis on security and surveillance also plays a pivotal role in the growth trajectory of the image recognition market. Governments and enterprises are investing heavily in advanced surveillance systems that employ facial recognition, object detection, and pattern recognition to enhance public safety and prevent crime. The integration of image recognition with existing security infrastructure enables real-time threat detection, automated monitoring, and improved law enforcement capabilities. Furthermore, regulatory mandates and compliance requirements in sectors such as banking, financial services, and insurance (BFSI) are driving the deployment of robust authentication and fraud prevention solutions powered by image recognition. These factors collectively contribute to the sustained growth and adoption of image recognition technologies across diverse industry verticals.

Regionally, North America continues to dominate the image recognition market, accounting for the largest revenue share in 2024, followed by Asia Pacific and Europe. The presence of leading technology providers, early adoption of AI-driven solutions, and substantial investments in research and development are key factors underpinning North America's leadership. Meanwhile, Asia Pacific is witnessing rapid growth due to increasing digitalization, expanding smartphone penetration, and rising investments in smart city initiatives. Europe remains a significant market, driven by strong demand in automotive, healthcare, and retail sectors. The Middle East & Africa and Latin America are also experiencing steady growth, albeit at a comparatively slower pace, as adoption of image recognition technologies accelerates in these regions.

Component Analysis

The image recognition market is se

Outsourced Data Labeling Market Outlook

The global outsourced data labeling market size was valued at approximately USD 1.6 billion in 2023 and is projected to reach around USD 10.2 billion by 2032, growing at a compound annual growth rate (CAGR) of 22.3% during the forecast period. This significant growth is driven by the increasing adoption of artificial intelligence and machine learning technologies across various industries, which has necessitated the need for high-quality annotated data to train these advanced systems.

One of the primary growth factors for the outsourced data labeling market is the burgeoning demand for AI-driven solutions in industries such as healthcare, automotive, and retail. As companies strive to leverage AI for enhancing operational efficiency, customer experience, and decision-making processes, the need for accurately labeled data sets has become paramount. This has led to a surge in demand for outsourced data labeling services, as organizations often lack the resources to manage data annotation internally.

Additionally, the proliferation of big data is another crucial factor propelling the market. The exponential increase in data generation from various sources, including social media, IoT devices, and digital transactions, has created a massive repository of data that needs to be processed and labeled for meaningful insights. Outsourced data labeling provides a viable solution for handling large volumes of data efficiently, enabling companies to focus on their core competencies while leveraging expert services for data annotation.

The rise of autonomous vehicles and advanced driver-assistance systems (ADAS) is also a significant contributor to the market’s growth. The automotive sector is heavily reliant on precise data labeling to train AI models for object detection, lane recognition, and other critical functionalities. Outsourcing these tasks to specialized vendors ensures high-quality annotations, speeds up the development process, and reduces the overall time-to-market for new technologies.

Regionally, North America is expected to hold a significant share of the outsourced data labeling market. This can be attributed to the presence of numerous tech giants and startups focusing on AI and machine learning in the region. Furthermore, the robust infrastructure, government support, and availability of skilled professionals make North America a favorable market for outsourced data labeling services. Asia Pacific is also anticipated to witness substantial growth due to the increasing adoption of AI technologies in countries like China, Japan, and India.

Data Type Analysis

The outsourced data labeling market is segmented by data type into text, image, video, and audio. Text data labeling is one of the most prevalent segments due to its wide application across various industries. Annotated text is essential for natural language processing (NLP) tasks such as sentiment analysis, chatbots, and machine translation. The increasing adoption of AI-driven customer service applications and sentiment analysis tools is driving the demand for outsourced text data labeling services.

Image data labeling is another critical segment, primarily driven by the requirements of computer vision applications. This includes facial recognition, object detection, and medical image analysis. The healthcare sector significantly benefits from image annotation as it aids in the diagnosis and treatment planning by providing accurately labeled medical images. As AI continues to revolutionize the healthcare industry, the demand for image data labeling is expected to rise substantially.

Video data labeling is gaining traction due to its application in autonomous vehicles, security surveillance, and entertainment. In the automotive industry, video annotation is crucial for developing self-driving vehicles, where labeled video data is used to train models for detecting obstacles, recognizing traffic signs, and predicting pedestrian movements. The growing investments in autonomous vehicle technology are expected to drive the demand for video data labeling services.

Audio data labeling is essential for speech recognition and voice-controlled applications. With the increasing popularity of virtual assistants like Amazon Alexa, Google Assistant, and Apple's Siri, the need for accurate

This Project consists of two datasets, both of aerial images and videos of dolphins, being taken by drones. The data was captured from few places (Italy and Israel coast lines).

The aim of the project is to examine automated dolphins detection and tracking from aerial surveys.

The project description, details and results are presented in the paper (link to the paper).

Each dataset was organized and set for a different phase of the project. Each dataset is located in a different zip file:

1. Detection - Detection.zip

2. Tracking - Tracking.zip

Further information about the datasets' content and annotation format is below.

• In aim to watch each file content, use the preview option, in addition a description appears later on this section.

Detection Dataset

This dataset contains 1125 aerial images, while an image can contain several dolphins.

The detection phase of the project is done using RetinaNet, supervised deep learning based algorithm, with the implementation of Keras RetinaNet. Therefore, the data was divided into three parts - Train, Validation and Test. The relations is 70%, 15%, 15% respectively.

The annotation format follows the requested format of that implementation (Keras RetinaNet). Each object, which is a dolphin, is annotated as a bounding box coordinates and a class. For this project, the dolphins were not distinguished into species, therefore, a dolphin object is annotated as a bounding box, and classified as a 'Dolphin'. Detection zip file includes:

A folder for each - Train, Validation and Test subsets, which includes the images

An annotations CSV file for each subset

A class mapping csv file (one for all the subsets).

*The annotation format is detailed in Annotation section.

Detection zip file content:

Detection |——————train_set (images) |——————train_set.csv |——————validation_set (images) |——————train_set.csv |——————test_set (images) |——————train_set.csv └——————class_mapping.csv

Tracking

This dataset contains 5 short videos (10-30 seconds), which were trimmed from a longer aerial videos, captured from a drone.

The tracking phase of the project is done using two metrics:

VIAME application, using the tracking feature

Re3: Real-Time Recurrent Regression Networks for Visual Tracking of Generic Objects, by Daniel Gordon. For this project, the author's Tensorflow implementation is being used

Both metrics demand the videos' frames sequence as an input. Therefore, the videos' frames were extracted. The first frame was annotated manually for initialization, and the algorithms track accordingly. Same as the Detection dataset, each frame can includes several objects (dolphins).

For annotation consistency, the videos' frames sequences were annotated similar to the Detection Dataset above, (details can be found in Annotation section). Each video's frames annotations separately. Therefore, Tracking zip file contains a folder for each video (5 folders in total), named after the video's file name.

Each video folder contains:

Frames sequence directory, which includes the extracted frames of the video

An annotations CSV file

A class mapping CSV file

The original video in MP4 format

The examined videos description and details are displayed in 'Videos Description.xlsx' file. Use the preview option for displaying its content.

Tracking zip file content:

Tracking |——————DJI_0195_trim_0015_0045 | └——————frames (images) | └——————annotations_DJI_0195_trim_0015_0045.csv | └——————class_mapping_DJI_0195_trim_0015_0045.csv | └——————DJI_0195_trim_0015_0045.MP4 |——————DJI_0395_trim_0010_0025 | └——————frames (images) | └——————annotations_DJI_0395_trim_0010_0025.csv | └——————class_mapping_DJI_0395_trim_0010_0025.csv | └——————DJI_0195_trim_0015_0045.MP4 |——————DJI_0395_trim_00140_00150 | └——————frames (images) | └——————annotations_DJI_0395_trim_00140_00150.csv | └——————class_mapping_DJI_0395_trim_00140_00150.csv | └——————DJI_0395_trim_00140_00150.MP4 |——————DJI_0395_trim_0055_0085 | └——————frames (images) | └——————annotations_DJI_0395_trim_0055_0085.csv | └——————class_mapping_DJI_0395_trim_0055_0085.csv | └——————DJI_0395_trim_0055_0085.MP4 └——————HighToLow_trim_0045_0070 └—————frames (images) └—————annotations_HighToLow_trim_0045_0070.csv └—————class_mapping_HighToLow_trim_0045_0070.csv └—————HighToLow_trim_0045_0070.MP4

Annotations format

Both datasets have similar annotation format which is described below. The data annotation format, of both datasets, follows the requested format of Keras RetinaNet Implementation, which was used for training in the Dolphins Detection phase of the project.

Each object (dolphin) is annotated by a bounding box left-top and right-bottom coordinates and a class. Each image or frame can includes several objects. All data was annotated using Labelbox application.

For each subset (Train, Validation and Test of Detection dataset, and each video of Tracking Dataset) there are two corresponded CSV files:

Annotations CSV file

Class mapping CSV file

Each line in the Annotations CSV file contains an annotation (bounding box) in an image or frame. The format of each line of the CSV annotation is:

path/to/image.jpg - a path to the image/frame

x1, y1 - image coordinates of the left upper corner of the bounding box

x2, y2 - image coordinates of the right bottom corner of the bounding box

class_name - class name of the annotated object

path/to/image.jpg,x1,y1,x2,y2,class_name

An example from train_set.csv:

.\train_set\1146_20170730101_ce1_sc_GOPR3047 103.jpg,506,644,599,681,Dolphin .\train_set\1146_20170730101_ce1_sc_GOPR3047 103.jpg,394,754,466,826,Dolphin .\train_set\1147_20170730101_ce1_sc_GOPR3047 104.jpg,613,699,682,781,Dolphin .\train_set\1147_20170730101_ce1_sc_GOPR3047 104.jpg,528,354,586,443,Dolphin .\train_set\1147_20170730101_ce1_sc_GOPR3047 104.jpg,633,250,723,307,Dolphin

This defines a dataset with 2 images:

1146_20170730101_ce1_sc_GOPR3047 103.jpg which contains 2 objects classified as 'Dolphin'

1146_20170730101_ce1_sc_GOPR3047 104.jpg which contains 3 objects classified as 'Dolphin'

Each line in the Class Mapping CSV file contains a mapping:

class_name,id

An example:

Dolphin,0

Snapshot img

AI Video Analytic Market Size 2025-2029

The AI video analytic market size is forecast to increase by USD 14 billion, at a CAGR of 31.7% between 2024 and 2029.

The market is experiencing significant growth, driven by the expanding demand for actionable intelligence beyond security applications. This trend is fueled by the increasing adoption of video analytics in various industries, including retail, healthcare, and transportation, to optimize operations, enhance customer experience, and improve safety. However, the market faces several challenges. Heightened privacy concerns demand robust data security and privacy measures to protect sensitive information.
Furthermore, a complex regulatory landscape, with varying data protection laws and regulations, necessitates compliance and adaptation to ensure business continuity. Companies seeking to capitalize on market opportunities and navigate challenges effectively must focus on developing innovative, efficient, and secure video analytics solutions while ensuring regulatory compliance. The proliferation of edge computing and hybrid architectures necessitates the development of more efficient and scalable video analytics solutions.

What will be the Size of the AI Video Analytic Market during the forecast period?

Explore in-depth regional segment analysis with market size data - historical 2019-2023 and forecasts 2025-2029 - in the full report.
Request Free Sample

The video analytic market continues to evolve, driven by advancements in artificial intelligence (AI) and machine learning technologies. Video metadata extraction and annotation are essential components of this market, enabling object tracking systems and video data annotation for video object detection. AI-powered surveillance and image recognition technology are at the heart of these applications, with neural network architecture and risk assessment scoring enhancing their capabilities. Customer behavior tracking and people counting systems leverage anomaly detection systems and motion detection sensors for security threat detection. Video analytics platforms and behavioral analytics tools provide valuable insights, while video analytics dashboards and activity recognition models facilitate real-time video processing and data visualization.

Edge computing solutions and pattern recognition software optimize video stream processing, with optical character recognition and license plate recognition further expanding the market's reach. Computer vision algorithms, real-time video processing, and data labeling services form the foundation of machine learning pipelines. Predictive analytics models and facial recognition software offer additional value, with traffic flow optimization and cloud-based video storage rounding out the market's offerings. For instance, a leading retailer implemented a video analytics solution that increased sales by 15% through optimized traffic flow and improved customer experience. The global video analytic market is expected to grow by over 17% annually, reflecting its continuous dynamism and evolving patterns.

How is this AI Video Analytic Industry segmented?

The AI video analytic industry research report provides comprehensive data (region-wise segment analysis), with forecasts and estimates in 'USD million' for the period 2025-2029, as well as historical data from 2019-2023 for the following segments.

Component

  Software
  Services


Deployment

  Cloud based
  On premises


Sector

  Large enterprises
  SMEs


Application

  Security and surveillance
  Operational efficiency
  Customer analytics
  Risk management
  Compliance


Geography

  North America

    US
    Canada


  Europe

    France
    Germany
    UK


  APAC

    China
    India
    Japan
    South Korea


  South America

    Brazil


  Rest of World (ROW)

By Component Insights

The Software segment is estimated to witness significant growth during the forecast period. The market is witnessing significant growth as businesses increasingly leverage advanced software solutions to extract valuable insights from video data. This software component, the market's intellectual backbone, incorporates intricate algorithms, machine learning models, and application platforms to interpret video streams in real-time. The market is marked by continuous innovation and versatile deployment options, such as on-premises, cloud, and edge-based solutions. Beyond basic motion detection, AI-driven advancements in this sector now offer granular object classification, attribute recognition, real-time behavioral analysis, and anomaly detection. Deep learning and neural networks power these capabilities, enhancing accuracy and minimizing false positives.

For instance, a leading retailer reported a 15% increase in sales due to improved customer behavior tracking and traffic flow optimization. Industry experts anticip

Chinese Chemical Safety Signs (CCSS)

This dataset is compiled as a benchmark for recognizing chemical safety signs from images. We provide both the dataset and the experimental results.

1. The Dataset

The complete dataset is contained in the folder ccss/data. The images include signs based on the Chinese standard "Safety Signs and their Application Guidelines" (GB 2894-2008) for safety signs in chemical environments. This standard, in turn, refers to the standards ISO 7010 (Graphical symbols – Safety Colours and Safety Signs – Safety signs used in workplaces and public areas), GB/T 10001 (Public Information Graphic Symbols for Signs), and GB 13495 (Fire Safety Signs)

1.1. Image Collection

We collect photos of commonly used chemical safety signs in chemical laboratories and chemistry teaching. For a discussion of the standards we base our collections, refer to the book "Talking about Hazardous Chemicals and Safety Signs" for common signs, and refer to the safety signs guidelines (GB 2894-2008).

• The shooting was mainly carried out in 6 locations, namely on the road, in a parking lot, construction walls, in a chemical laboratory, outside near big machines, and inside the factory and corridor.
• Shooting scale: Images in which the signs appear in small, medium and large scales were taken for each location by shooting photos from different distances.
• Shooting light: good lighting conditions and poor lighting conditions were investigated.
• Part of the images contain multiple targets and the other part contains only single signs.

Under all conditions, a total of 4650 photos were taken in the original data. These were expanded to 27,900 photos were via data enhancement. All images are located in folder ccss/data/JPEGImages.

The file ccss/data/features/enhanced_data_to_original_data.csv provides a mapping between the enhanced image name and the corresponding original image.

1.2. Annotation and Labelimg

We use Labelimg as labeling tool, which, in turn, uses the PASCAL-VOC labelimg format. The annotation is stored in the folder ccss/data/Annotations.

Faster R-CNN and SSD are two algorithms that use this format. When training YOLOv5, you can run trans_voc2yolo.py to convert the XML file in PASCAL-VOC format to a txt file.

We provide further meta-information about the dataset in form of a CSV file features.csv which notes, for each image, which other features it has (lighting conditions, scale, multiplicity, etc.). We apply the COCO standard for deciding whether a target is small, medium, or large in size.

1.3. Dataset Features

As stated above, the images have been shot under different conditions. We provide all the feature information in folder ccss/data/features. For each feature, there is a separate list of file names in that folder. The file ccss/data/features/features_on_original_data.csv is a CSV file which notes all the features of each original image.

1.4. Dataset Division

The data set is fixedly divided into 7:3 training set and test set. You can find the corresponding image names in the files ccss/data/training_data_file_names.txt and ccss/data/test_data_file_names.txt.

2. Baseline Experiments

We provide baseline results with five models, namely Faster R-CNN (R), Faster R-CNN (M), SSD, YOLOv3-spp, and YOLOv5. All code and results is given in folder ccss/experiment.

2.2. Environment and Configuration:

• Single Intel Core i7-8700 CPU
• NVIDIA GTX1060 GPU
• 16 GB of RAM
• Python: 3.8.10
• pytorch: 1.9.0
• pycocotools: pycocotools-win
• Visual Studio 2017
• Windows 10

2.3. Applied Models

The source codes and results of the applied models is given in folder ccss/experiment with sub-folders corresponding to the model names.

2.3.1. Faster R-CNN

• Faster R-CNN (R) has the backbone resnet50+fpn.
  • we downloaded the pre-training weights from https://download.pytorch.org/models/fasterrcnn_resnet50_fpn_coco-258fb6c6.pth
  • we modify the type information of the JSON file to match our application.
  • run train_res50_fpn.py
  • finally, the weights trained by the training set.
  The Faster R-CNN (R) source code used in our experiment is given in folder ccss/experiment/sources/faster_rcnn (R). The weights of the fully-trained Faster R-CNN (R) model are stored in file ccss/experiment/trained_models/faster_rcnn (R).pth. The performance measurements of Faster R-CNN (R) are stored in folder ccss/experiment/performance_indicators/faster_rcnn (R).
• Faster R-CNN (M) has the backbone mobilenetv2.
  • backbone: MobileNetV2.
  • we modify the type information of the JSON file to match our application.
  • run train_mobilenetv2.py
  • finally, the weights trained by the training set.
  The Faster R-CNN (M) source code used in our experiment is given in folder ccss/experiment/sources/faster_rcnn (M). The weights of the fully-trained Faster R-CNN (M) model are stored in file ccss/experiment/trained_models/faster_rcnn (M).pth. The performance measurements of Faster R-CNN (M) are stored in folder ccss/experiment/performance_indicators/faster_rcnn (M).

2.3.2. SSD

• backbone: resnet50
• we downloaded pre-training weights from https://download.pytorch.org/models/resnet50-19c8e357.pth
• the same training method as Faster R-CNN is applied.

The SSD source code used in our experiment is given in folder ccss/experiment/sources/ssd. The weights of the fully-trained SSD model are stored in file ccss/experiment/trained_models/ssd.pth. The performance measurements of SSD are stored in folder ccss/experiment/performance_indicators/ssd.

2.3.3. YOLOv3-spp

• backbone: DarkNet53
• we modified the type information of the XML file to match our application
• run trans_voc2yolo.py to convert the XML file in VOC format to a txt file.
• the weights used are: yolov3-spp-ultralytics-608.pt.

The YOLOv3-spp source code used in our experiment is given in folder ccss/experiment/sources/yolov3-spp. The weights of the fully-trained YOLOv3-spp model are stored in file ccss/experiment/trained_models/yolov3-spp.pt. The performance measurements of YOLOv3-spp are stored in folder ccss/experiment/performance_indicators/yolov3-spp.

2.3.4. YOLOv5

• backbone: CSP_DarkNet
• we modified the type information of the XML file to match our application
• run trans_voc2yolo.py to convert the XML file in VOC format to a txt file.
• the weights used are: yolov5s.

The YOLOv5 source code used in our experiment is given in folder ccss/experiment/sources/yolov5. The weights of the fully-trained YOLOv5 model are stored in file ccss/experiment/trained_models/yolov5.pt. The performance measurements of YOLOv5 are stored in folder ccss/experiment/performance_indicators/yolov5.

2.4. Evaluation

The computed evaluation metrics as well as the code needed to compute them from our dataset are provided in the folder ccss/experiment/performance_indicators. They are provided over the complete test st as well as separately for the image features (over the test set).

3. Code Sources

1. Faster R-CNN (R and M)
   • https://github.com/WZMIAOMIAO/deep-learning-for-image-processing/tree/master/pytorch_object_detection/faster_rcnn
   • official code: https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py
2. SSD
   • https://github.com/WZMIAOMIAO/deep-learning-for-image-processing/tree/master/pytorch_object_detection/ssd
   • official code: https://github.com/pytorch/vision/blob/main/torchvision/models/detection/ssd.py
3. YOLOv3-spp
   • https://github.com/WZMIAOMIAO/deep-learning-for-image-processing/tree/master/pytorch_object_detection/yolov3-spp
4. YOLOv5
   • https://github.com/ultralytics/yolov5

We are particularly thankful to the author of the GitHub repository WZMIAOMIAO/deep-learning-for-image-processing (with whom we are not affiliated). Their instructive videos and codes were most helpful during our work. In