Scribd
Upload
14 views20 pages

PBBML L10

The document discusses image classification AI, detailing the process of assigning labels to images based on their content, including steps like pre-processing, object detection, and classification. It highlights the Wildlife Insights project, which uses AI and camera traps for wildlife monitoring and conservation, and describes the technical approach and hardware/software requirements for building an image classification app. Additionally, it introduces Pandas, a Python library for data manipulation and analysis, emphasizing its functionalities for data cleaning, aggregation, and transformation.

Uploaded by

Ooi Wei Ying
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
14 views20 pages

PBBML L10

The document discusses image classification AI, detailing the process of assigning labels to images based on their content, including steps like pre-processing, object detection, and classification. It highlights the Wildlife Insights project, which uses AI and camera traps for wildlife monitoring and conservation, and describes the technical approach and hardware/software requirements for building an image classification app. Additionally, it introduces Pandas, a Python library for data manipulation and analysis, emphasizing its functionalities for data cleaning, aggregation, and transformation.

Uploaded by

Ooi Wei Ying
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 20

Python Based

Basic Machine Learning


Dr. –Ing. Youjin Kim
Video2
인문대학2호관 331
1:00pm ~
Lec 10
• Image Classification AI
• Google & Wildlifeinsight
• Pandas Practice
Image Classification AI
• image classification is a process of classifying and assigning labels to
groupings of images or vectors within an image based on certain criteria.
A label can be assigned based on one or more criteria.
• Image classification assigns a label to an entire image based on the
objects or features present within it. For example, an image might be
classified as "cat," "dog," or "car," depending on what is depicted.

https://youtu.be/iYmMyKE6FN4?si=fYYSPmTO7fh9If5g
Image Classification AI
• image classification is a process of classifying and assigning labels to
groupings of images or vectors within an image based on certain criteria.
A label can be assigned based on one or more criteria.
• Image classification assigns a label to an entire image based on the
objects or features present within it. For example, an image might be
classified as "cat," "dog," or "car," depending on what is depicted.

https://youtu.be/iYmMyKE6FN4?si=fYYSPmTO7fh9If5g
Image Classification AI
• Image classification assigns a label to an entire image based on the
objects or features present within it.
• For example, an image might be classified as "cat," "dog," or "car,"
depending on what is depicted.
(Compare this to Image Generative AI)

https://www.superannotate.com/blog/image-classification-basics
Image Classification Steps
1. Pre-processing: This step involves cleaning and preparing the
image data for processing by the AI model. This includes tasks
such as removing noise, resizing images, and normalizing pixel
values.
2. Object Detection: Identifying and locating objects within the
image. This is done using algorithms that can detect edges,
corners, and other features in the image.

https://youtu.be/iYmMyKE6FN4?si=fYYSPmTO7fh9If5g
Image Classification Steps
3. Object Recognition and Training: It labels the detected objects in
the image. This can be done manually or automatically using ML
algorithms. Once the objects are labeled, the AI model is trained on a
large dataset of labeled images.
4. Object Classification: Using the trained AI model to classify new
images. The model analyzes the features of the new image and
assigns it to one of the predefined classes.

https://youtu.be/iYmMyKE6FN4?si=fYYSPmTO7fh9If5g
Single-label classification
vs Multi-label classification

https://youtu.be/iYmMyKE6FN4?si=fYYSPmTO7fh9If5g
https://youtu.be/hUzODH3uGg0?si=hCnr3DYlAFLj58HS

Wildlife Image Classification


• WildlifeInsights.org uses AI and
camera traps to streamline wildlife
monitoring for conservation.
• Camera traps, equipped with motion
sensors, automatically capture images
when animals pass by.
Image: Google Cloud Tech, Recovering Global Wildlife Populations using ML

• These images are then uploaded to the


Wildlife Insights platform, where AI
models analyze them to identify species,
filter out blank images, and detect
possible threats like poachers.

Image: Wildlife Insight, https://youtu.be/qKgRbkCkRFY?si=OqzSZo0HuaQm-Ncm


Wildlife Image Classification
• Purpose of Camera Traps: Used to
monitor wildlife populations non-
invasively with motion-sensitive cameras.
• Challenge: Reviewing millions of images
and categorizing species while detecting
poachers. Image: Google Cloud Tech, Recovering Global Wildlife Populations using ML

• Solution Overview: Demonstrates the


creation of an image classification app
using machine learning for automating
data analysis.

Image: Wildlife Insight, https://youtu.be/qKgRbkCkRFY?si=OqzSZo0HuaQm-Ncm


Project Highlight - Wildlife Insights
Collaboration with Google:
• Google partnered with conservation organizations to create a
classifier and web app for camera trap data.
• Allows data sharing across organizations to study diverse
species.
• Hosted on WildlifeInsights.org, this project addresses wildlife
decline, with populations reduced by 68% since the 1970s.
Technical Approach:
• Data Sharing Challenge: Many institutions keep data isolated; this project
centralizes and processes it on Google Cloud.
• AI Application: Models predict and identify species within images.
• Hardware: Camera traps linked to microcontrollers (e.g., Raspberry Pi)
perform on-site classification and image uploads.

Image: Google Cloud Tech, Recovering Global Wildlife Populations using ML


Building the APP
Hardware Requirements:
• Camera traps connected to microcontrollers
upload labeled photos; remote transfers rely on
Wi-Fi or periodic syncs.
• Automation Tip: Use scripts with Cloud IoT Core https://www.raspberrypi.com/
for remote device updates.

Image: Google Cloud Tech, Recovering Global Wildlife Populations uisng ML


Software Requirements:
• Dataflow: Processes large volumes of data,
creating a balanced dataset in BigQuery.
• AI Platform: Develops and deploys the model using
AutoML for image classification.
• Cost Estimate: Under $30 per month for basic data
processing and storage.

Image: Google Cloud Tech, Recovering Global Wildlife Populations uising ML


Model Development and Usage
• Training Model: Dataflow uploads necessary
metadata and images.
• Classification: AI Platform processes and identifies
species.
Deployment: Microcontrollers classify and upload
images for real-time conservation data.

Image: Google Cloud Tech, Recovering Global Wildlife Populations uising ML


Project Highlight - Wildlife Insights

Image: Google Cloud Tech, Recovering Global Wildlife Populations uising ML


Pandas
• Pandas is an open-source Python library that is widely used for
data manipulation and analysis, especially in data science and
machine learning.
• It provides data structures and functions designed to work with
structured data, like data tables or time series, making it easier
to clean, transform, and analyze data efficiently.
Pandas
Data Manipulation and Cleaning
• Filtering and Selecting Data: You can filter data based on
conditions, select specific columns, or retrieve rows that meet
particular criteria.
• Handling Missing Data: Functions like dropna() and fillna() allow
you to handle missing data by removing or replacing it with
default values.
• Sorting and Ranking: Pandas provides functions like
sort_values() to sort data by specific columns.
• Transforming Data: Using methods like apply(), you can apply
custom functions to modify data within your DataFrame.
Pandas
Data Aggregation and Grouping
• You can use the groupby() function to group data by one or more
columns and apply aggregation functions like sum(), mean(),
count(), etc., which is helpful for summarizing data by categories.
Pandas
Data Aggregation and Grouping
• Practice with train.csv and ED.csv files

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy