DHANALAKSHMI SRINIVASAN COLLEGE OF ENGINEERING (AUTONOMOUS) G&G COIMBATORE - 641105 | www.dsce.ac.in (Approved by AICTE, New Delhi & Affiliated to Anna University, Chennai) NH- 47, Palakkad Main Road, Navakkarai Po, Coimbatore — 641 105. DEPARTMENT OF ARTIFICIAL INTELLIGENCE AND DATA SCIENCE RECORD NOTEBOOK Name: Register No: ‘ Subject Code/Title: Year/Semester: Academic Year:DHANALAKSHMI SRINIVASAN COLLEGE OF ENGINEERING (AUTONOMOUS) COIMBATORE - 641105 | www.dsce.ac.in (Approved by AICTE, New Delhi & Affiliated to Anna University, Chennai) NH- 47, Palakkad Main Road, Navakkarai Po, Coimbatore — 641 105. DEPARTMENT OF ARTIFICIAL INTELLIGENCE AND DATA SCIENCE CERTIFICATE This is to certify that Mr./Ms. Reg. No. a of, Semester _Year B.Tech. (Artificial Intelligence and Data Science) has completed his/her practical work in the. laboratory during the academi¢ year 2024-25. Faculty In-charge Head of the Department Submitted for the University practical examination held on : INTERNAL EXAMINER EXTERNAL EXAMINERINDEX S.No. dimly Name of the Experiment No. Mark Staff Sign.Ex.no:1 CONSTRUCT HUFFMAN CODES FOR GIVEN Date: SYMBOL PROBABILITIES Aim: To implement the Huffman coding algorithm for a given set of symbol probabilities and construct efficient prefix codes. Algorithm: 1. Input: A list of symbols with their corresponding probabilities or frequencies. . Create a priority queue (min-heap) containing all symbols. . While there is more than one node in the queue: Remove two nodes with the lowest probabilities. Create a new internal node with probability equal to the sum of two. Assign 0 and 1 as edges to the two child nodes. Insert the new node back into the queue. 4, Repeat until the queue contains only one node — the root of the Huffman tree. 5.Traverse the tree to assign binary codes to cach symbol: eee enn Source Code: import heapq class Node: def _init_(self, symbol, prob): self.symbol = symbol self.prob = prob self.left = None self.right = None def _It_(self, other): return self.prob < other.prob def huffman_codes(symbols_probs): heap = [Node(sym, prob) for sym, prob in symbols_probs.items()] heapg.heapify(heap) while len(heap) > 1: left = heapg.heappop(heap) right = heapg.heappop(heap) new_node = Node(None, left.prob + right.prob)new_node.left = left new_node.right = right heapq.heappush(heap, new_node) root = heap[0] codes = {} def generate_codes(node, code=""): if node is not None: if node.symbol is not None: codes[node.symbol] = code generate_codes(node.left, code + "0") generate_codes(node.right, code + "1") generate_codes(root) return codes Output: Symbol Huffman Code A 0 in 101 1001 1000 Ee oo ow Result: The Hufliman coding algorithm successfully generated variable-length, prefix-fiee codes. Symbols with higher probabilities have shorter codes, minimizing the total weighted path length.Ex.no:2 ENCODE RUN LENGTHS WITH FIXED-LENGTH Date: CODE USING PYTHON Aim: To implement run-length encoding (RLE) using fixed-length codes to compress a string by encoding repeated characters. Algorithm: 1. Inputa string of characters. 2. Traverse the string while counting consecutive occurrences of each character. 3, Store each character followed by ifs run-length count, 4. Use a fixed-length binary representation for counts (e.g., 4-bit for max count 15). 5. Output the encoded string in character + binary count format. Source Code: def fixed_length_binary(n, bits=4): return format(n, £0 {bits}b’) def run_length_encode_fixed(s): encoded = "" i=0 while i /dev/null then echo "ImageMagick is not installed. Please install it using 'sudo apt install imagemagick" exit | fi # Loop through image files (add more extensions if needed) for img in *.png *.bmp *.gif *.tiff; do if [-f"Simg" J; then filename="$ {img%.*}" convert "Simg" "Sfilename.jpg” echo "Simg converted to Sfilename.jpg" fi done Output: If the current directory contains the following files: image L.pngphoto2.bmp icon3.gif After running the script: image! jpg photo2,jpg icon3.jpg, will be created in the same directory. Result: The shell script successfully converted all image files in the current directory to JPEG format using ImageMagick.Ex.no:6 WRITE A PROGRAM TO SPLIT IMAGES FROM A Date: VIDEO WITHOUT USING ANY PRIMITIVES Aim: To extract frames (images) from a video file manually, avoiding high-level primitives, and save them as individual image files. Algorithm: 1. Import necessary libraries (cv2, 0s). 2. Open the video file using v2. VideoCapture. 3. Retrieve the total frame count and frame rate. 4. Loop through the frames by manually setting the frame position using set() and extracting each frame using read(). 5. Save each frame using ev2.imwrite() into an output folder. 6. Close the video capture. Source Code: import ev2 import os def split_video_to_images(video_path, output_folder): # Create output folder if it doesn't exist if not os.path.exists(output_folder): os.makedirs(output_folder) cap = cv2.VideoCapture(video_path) if not cap.isOpened(): print("Error: Could not open video.") return total_frames = int(cap.get(ev2.CAP_PROP_FRAME_COUNT)) print(f'Total frames: {total_frames}") for frame_num in range(total_frames): cap.set(ev2.CAP_PROP_POS_FRAMES, frame_num) # Manually set frame position success, frame = cap.read() # Read frame if not success: print("Error reading frame {frame_num}") continuefilename = os.path,join(output_folder, f"frame_{frame_num:04d} jpg") cv2.imwrite(filename, frame) cap.release() print(""Frames successfully extracted.") # Example usage video_file = "sample_video.mp4" # Replace with your video file path output_dir = "extracted_frames" split_video_to_images(video_file, output_dir) Output: The program will output image files such as: extracted_frames/frame_0000.jpg extracted_frames/frame_0001.jpg Each image corresponds to a frame from the video. Result: The program successfully extracts all frames from the video without using high-level primitives like automatic looped reading. It manually navigates frame-by-frame and saves them to disk, providing low-level control over video processing.Ex.no:7 CREATE A PHOTO ALBUM OF A TRIP BY Date: APPLYING APPROPRIATE IMAGE DIMENSIONS, AND FORMAT Aim: To develop a photo album of a trip by resizing and formatting images appropriately using Python, ensuring uniformity in display and compatibility across digital platforms. Algorithm: 1. Import Required Libraries: Use PIL (Python Imaging Library) for image processing. 2. Set Target Dimensions and Format: Decide the final image size (e.g., 800x600 pixels) and format (¢.g., JPEG or PNG). 3. Load Images: Access all images from a selected folder. 4, Resize and Format Images: Open each image. Resize it to the target dimensions. Convert and save it in the desired format. 5. Create Album Layout: Use a basic HTML template or Python script to generate a layout. 6. Display Album: View the album using a web browser or simple GUI. Source Code: from PIL import Image import os # Define input and output folders input_folder = 'trip_images’ output_folder = 'album_images’ os.makedirs(output_folder, exist_ok=True) # Desired image size and format target_size = (800, 600) target_format = 'JPEG' # Process images for filename in os.listdir(input_folder): if filename.lower().endswith((',jpe’, ‘png’, jpeg’): img_path = os,path,join(input_folder, filename) img = Image.open(img_path) img_resized = img resize(target_size)output_path = os.path,join(output_folder, os.path.splitext(filename)[0] + ips) img_resized.save(output_path, target_format) print(""Images resized and formatted successfully." Output: All input images resized to 800x600 pixels. Converted and saved in JPEG format inside the album_images folder. These images are now consistent in size and format, suitable for a web or desktop album, Result: The photo album was successfully created with resized and uniformly formatted images, demonstrating how Python can be used for basic digital image processing tasks. The output album is easy to view and share due to its consistent format.Ex.no:8 IDENTIFYING THE POPULARITY OF CONTENT Date: RETRIEVAL FROM A MEDIA SERVER Aim: To write a program that analyzes content request logs to identify the popularity of each content item retrieved from a media server. Algorithm: 1. Start 2. Create a log of content requests (¢.g., video names or content IDs). 3. Initialize a dictionary to store the count of requests for each content. 4. Loop through each request in the log: If the content is already in the dictionary, increment its count. Else, add it to the dictionary with count = 1. 5. Sort the dictionary based on the frequency of access in descending order. 6. Display the content along with their request counts. 7. Stop Source Code: # content_popularity.py. def identify_popularity(request_logs): popularity dict = {} # Count the number of times each content is requested for content in request_logs: if content in popularity_dict: popularity_dict[{content] += 1 else: popularity _dict[content] = 1 # Sort contents based on number of requests (popularity) sorted_popularity = sorted(popularity_dict.items(), key=lambda x: x[1], reverse=True) print("Content Popularity from Media Server:\n") print(""{:<20} {}".format("Content", "Number of Requests")) print("-" * 35) for content, count in sorted_popularity:print(" {:<20} {}".format(content, count)) # Sample log data (you can replace this with actual logs) media_requests = [ "movie1.mp4", "song].mp3", "moviel mp4", "movie2.mp4”, "song2.mp3", "movie1.mp4", "song|.mp3", "movie3.mp4", "movie2.mp4", "song 1.mp3" ] identify_popularity(media_requests) Output: Content Popularity from Media Server: Content Number of Requests moviel.mp4 3 song].mp3 3 movie2.mp4 2 song2.mp3 1 movie3.mp4 1 Result: The Python program successfully identified the popularity of content retrieved from the media server by counting the number of requests for each content and displaying them in descending order of popularity.Ex.no:9 ENSURING DATA AVAILABILITY IN DISKS USING. Date: STRIP-BASED METHOD Aim: To implement a striping method for data storage across multiple disks, ensuring data availability and improved performance by distributing data in strips across disks. This method is commonly used in RAID (Redundant Array of Independent Disks), especially RAID 0 (striping without parity) or RAID 5 (striping with parity). Algorithm: 1. Divide the data into blocks. 2. Distribute (strip) the blocks across multiple disks in a round-robin fashion. 3. Calculate parity for each set of blocks and store it on one of the disks in a rotating pattern. 4, During failure, the missing data block can be reconstructed using parity and the remaining data. Source Code: def calculate_parity(blocks): # XOR all data blocks to get the parity parity = 0 for block in blocks: parity “= block return parity with_parity(data_blocks, num_disks): for iin range(0, len(data_blocks), num_disks - 1): stripe_data = data_blocks[i:i + (num_disks - 1)] while len(stripe_data) <(num_disks - 1): stripe_data.append(0) # Padding if not enough data parity = calculate_parity(stripe_data) stripe ‘stripe_data) parity_position = (i // (num_disks - 1)) % num_disksstripe.insert(parity_position, parity) stripes.append{(stripe) return stripes def display_stripes(stripes): for i, stripe in enumerate(stripes): print(f"Stripe {i+1}: {stripe}") # Example usage data_blocks = [10, 20, 30, 40, 50, 60, 70] # Simulated data num_disks = 4 stripes = striping_with_parity(data_blocks, num_disks) display_stripes(stripes) Output: Stripe 1: [60, 10, 20, 30] Stripe 2: [40, 70, 50, 60] Explanation: In Stripe 1, the parity (60) is stored in disk 0. In Stripe 2, the parity (40) is stored in disk 1, rotating parity disk. Result: The code demonstrates how striping with parity improves data availability across multiple disks. Even if one disk fails, the lost data can be reconstructed using parity and surviving data. This is foundational in RAID 5 implementations for fault tolerance in multimedia storage systems.Ex.no:10 PROGRAM FOR SHEDULING REQUESTS FOR Date: DATA STREAMS Aim: To write a shell script that converts all image files in the current directory to JPEG format. Algorithm: 1. Start the script. 2. Loop through all image files in the current directory (e. GIF). 3. Use the convert command (from ImageMagick) to convert each image to JPEG format. 4, Save the output with a .jpg extension. 5. End the script. PNG, BMP, Source Code: #\/bin/bash # Check if ImageMagick is installed if ! command -v convert &> /dev/null then echo "ImageMagick is not installed, Please install it using 'sudo apt install imagemagick" exit 1 fi # Loop through image files (add more extensions if needed) for img in *.png *.bmp *.gif *.tiff; do if [-f"Simg" ]; then filename="$ {img%.*}" convert "Simg" "Sfilename,jpg" echo "Simg converted to $filename.jpg" fi done Output: If the current directory contains the following files: image |.pngphoto2.bmp icon3.gif After running the script: imagel jpg photo2.jpg icon3.jpg will be created in the same directory. Result: The shell script successfully converted all image files in the current directory to JPEG format using ImageMagick.